Flight Without Power

WITHOUT
POWER
SB
K8
PILOT'S VIEW WHEN TAKING OFF AT ELMIRA
7" Loom is
FLIGHT
WITHOUT POWER
The Art of Gliding and Soaring
By
LEWIN B. BARRINGER
Glider Specialist, Air Staff
Headquarters of the Army Air Forces
PITMAN PUBLISHING CORPORATION

NEW YORK
CHICAGO
COPYRIGHT, 1940
BY
PITMAN PUBLISHING CORPORATION
All rights reserved. No part of tfiis book

may be reproduced in any form without
the written permission of the publisher.
Revised Edition, 1942

Reprinted, Sept. 1942
Reprinted, Apr. 1943
ASSOCIATED COM TAN IKS
SIR ISAAC PITMAN & SONS, LTD.

Bath London Melbourne Johannesburg Singapore
SIR ISAAC PITMAN & SONS (CANADA), LTD.
381-383 Church Street, Toronto
Advisory Editor
PROFESSOR ALEXANDER KI.EMIN
DANIEL GUGGENHEIM SCHOOL OF AERONAIJIICS
COLLEGE OF ENGINEERING, NEW YORK UNIVERSITY
PRINTED IN Till', UNlll'.D SI A IKS OF AMERICA
DEDICATED TO THE MEMORY
OF
Warren E. Eaton
whose unselfish leadership did so much
to promote gliding and soaring in the
United States.
FOREWORD
During the two years after my first taste of motorlcss flying in

1930, I operated the Wings Gliding School near Philadelphia which
trained over sixty students largely by methods that my associates
and I had to devise due to lack of any adequate book or manual
on the subject. Following my introduction to high performance
soaring by my friend Richard du Pont in the summer of 1934, I
frequently encountered conditions and had experiences in flight
the correct explanation of which also could not be found in any
book yet published.
Many times during the two years as editor of Soaring and man
ager of The Soaring Society of America I felt an increasing need
for an up-to-date, authoritative book covering all phases of motorless flying. Although several excellent books have been written,
soaring technique and sailplane design have advanced so rapidly in
recent years that these are now inadequate. An indication of this
progress is the fact that while making several thousand flights at
Wings Field some years ago we never suspected possibilities of soar
ing over this field which is situated in level country. Now the mem
bers of a soaring club are being taught not only how to win their
"C" licenses but also how to go on to high performance soaring on
thermal upcurrents in this same part of the country. From being
restricted to a few isolated sites in the mountains, soaring can now
be done from large fields situated nearly anywhere in the country.
So it was that when approached to write this book I well knew
the great need for such a work but also realized that to be truly
authoritative certain chapters should be written by recognized ex
perts in their particular fields. I wish to express my appreciation for
the invaluable co-operation of my co-authors. H. Randers-Pehrson
is a recognized authority in the historical field and a member of
the staff of the Division of Aeronautics of the Library of Congress,
Washington, D. C. Paul and Ernest Schweizer of Elmira, N. Y.,
are leading glider designers and builders who pioneered all-metal
construction in this field. Karl O. Lange, an eminent meteor
ologist connected with the Blue Hill Meteorological Observatory
of Harvard University, is also a former soaring pilot, Contest
vii
viii
FOREWORD
Director for The Soaring Society of America and designer of one

of the first successful radio-meteorgraphs. Charles O. Colvin is a
well-known expert in the field of aircraft instruments.
In expressing my thanks and gratitude to those who have assisted
me I wish particularly to mention Milton Stoughton for his initial
work on the aerodynamics of gliding flight; Fred C. Barnes, John
Robinson, Gilbert Walters, Gustave Scheurer, and Amos Wood for
their information on the soaring sites at Tejon Ranch, Torrey
Pines Mesa, Altamount Pass, Schley Field, and Steptoe Butte, re
spectively; Jay Buxton for help on technical details of launching
methods used in California; A. Ivanoff, British "Silver C" pilot, for
his suggestion that English data and statistics be included; The
Soaring Society of America for permission to reprint certain passages
published in Soaring; and all those who helped so much by supply
ing photographs for illustrations.
It is my sincere hope that this book may help to promote the
soaring movement.
LEWIN B. BARRINGER
Llewellyn Park
Orange, N. J.
FOREWORD TO THE REVISED EDITION

Increased public interest in motorless flight has exhausted the
first edition of this book. In this revised edition we have eliminated
three chapters: Soaring Sites, Clubs, and the Future of Gliding and
Soaring. A new chapter on Transport Gliders has been added and
a revised list of gliding clubs and schools is now in the Appendix.
My gratitude and thanks are extended to August Raspet, physicist
with the U. S. Geological Survey, for the valuable, new material on
variometers, the explosive release and the aid to auto-pulley launch
ing. The remaining chapters are largely unchanged except for a few
additional items of value brought out by recent experience.
All the information necessary for the student of motorless flight
is in this edition.
In the foreword to the first edition the hope was expressed that
the book might help promote the soaring movement. I now express
the more serious and important hope that this edition will help
motorless flying to take its proper place in the national effort to
make us supreme in the air for the defense of democracy.
LEWIN B. BARRINGER
Washington, D. C.
CONTENTS
PAGE
FOREWORD
...............
FOREWORD TO THE REVISED EDITION
vii
........
viii
HISTORY OF MOTORLESS FLIGHT, by N. H. RandersPehrson

.............
AERODYNAMICS, by Paul Schweizer, Assisted by Milton

Stoughton and Ernest Schweizer ......
15
III
TYPES OF GLIDERS, by Lewin B. Barringer ....
33
IV
DESIGN, CONSTRUCTION AND MAINTENANCE, by Paul and

Ernest Schweizer ...........
40
LAUNCHING METHODS, by Lewin B. Barringer
...
84
SOARING METEOROLOGY, by Karl O. Lange .
...in
CHAPTER
I
II
V
VI
VII
VIII
INSTRUMENTS, by Charles H. Colvin and August Raspet
145
FLIGHT TRAINING, by Lewin B. Barringer
164
IX
SOARING TECHNIQUE, by Lewin B. Barringer .
175
TRANSPORT GLIDERS, by Lewin B. Barringer .
196
APPENDIX. Gliding and Soaring LicensesRecordsLists of

"Silver C" and "Golden C" Pilots-Gliding Clubs and
Schools in the United StatesBritish Gliding Clubs
. 201
BIBLIOGRAPHY
..............216
GLOSSARY ................
217
INDEX
219
................
ix
CHAPTER I
HISTORY OF MOTORLESS FLIGHT

By N. H. Randers-Pehrson
I HE EARLY HISTORY OF AVIATION is a history of failures

and frustrations because, not realizing the possibilities of gliding
and soaring, men tried to fly by muscular power and spent their
time and ingenuity on futile wing-flapping devices. Later when this
method had been discredited, practical progress was retarded by
the false belief that an engine is indispensable for human flight.
The foundation for scientific study of aviation was laid by Sir
George Cayley who in the beginning of the nineteenth century
made the first attempts to explain mathematically the principles of
flight. He also experimented with glider models, and built a large
machine which, although provided with some kind of propelling
mechanism, was tried in gliding flight. When launched from a hill
top without a pilot, it would sail with perfect balance to the plain
below, and when a person ran with it against the wind, it would
sometimes carry him a few yards. Probably Cayley had the knowl
edge and ability to build and operate a man-carrying glider, but
unfortunately he had his mind set upon the need for an engine and
so missed the opportunity of becoming the father of motorless
flight. He was followed by a number of gifted and enthusiastic
workers, mostly in England and France, who made valuable theo
retical contributions but achieved no practical results because they
also were striving for power flight. Among the few who favored
motorless flight was Captain Le Bris, a French sailor who used the
albatross as a pattern for a glider. With this he was towed into the
air like a kite in 1855, and again with a second Albatross in 1867,
but mishaps and lack of funds brought the experiments to an end.
Le Bris' glider had streamline form and a large aspect ratio; the
angle of incidence of the wings could be varied in flight. There was
also Louis Pierre Mouillard, a Frenchman who lived in Algeria and
who spent all his spare time studying those masters of soaring
flight, the great vultures. Mouillard described his observations in a
book entitled L'Empire de I'Air, a work which furnished much
inspiration to others, but his own attempts to make wings for glid
ing were crude and ineffectual.
FLIGHT WITHOUT POWER
Otto Lilienthal was the first man to learn the art of flight from
the birds, to practice the art himself, and to give it to humanity.
From boyhood he and his brother Gustav watched the birds, espe
cially the storks, so abundant near their home in Pomerania. After
years of study and aerodynamic experiments he produced the best
treatise on the theory of flight that had yet appeared. The next step
was to learn to fly. Lack of a suitable engine had been a stumbling
block for other inventors, but Lilienthal concluded that an engine
was not necessary or even desirable for pioneer flying. In 1891 he
made his first glider of peeled willow rods covered with waxed
fabric; it was shaped like a pair of broad bird's wings and had fixed
horizontal and vertical tail surfaces. During the next five years he
built several hang gliders, both monoplanes and biplanes. To oper
ate these, the pilot stood in the middle of the apparatus, thrusting
his arms through padded openings in the frame, so that the weight
in flight rested on his elbows. He maintained his balance in the air
by moving his body or swinging his legs to shift the center of gravity
and keep it directly under the center of lift. This required great
acrobatic ability, but by persistent practice Lilienthal mastered
the art and became so skillful that he could fly even in strong
winds. He was, however, aware of the need for a better system of
control, and on one of his last gliders used a movable elevator.
Lilienthal made his first flights from a springboard in his garden,
then from low hills in the neighborhood, and finally he built an
artificial hill 50 feet high, with a shed for the gliders in the top. He
also found a suitable gliding site at Rhinow, where there are low
mountains covered with heath and grass. Here his flights became
longer, up to 900 feet, and he succeeded in making turns, sometimes
of almost 180. It was at Rhinow on August 9, 1896, that he lost his
balance in the air, fell and was fatally injured. It was Lilienthal's
hope that others would take up gliding as a sport. Therefore he
published detailed reports of his activities and did not object to the
hundreds of spectators who flocked to his flying hill every Sunday.
The following he had wished for was slow in coming, but his work
was continued by a few capable disciples, notably Pilcher in Eng
land, Ferber in France, and Chanute, Herring and the Wright
brothers in America.
Percy S. Pilcher built several gliders with slight improvements
over Lilienthal's; he launched himself by running downhill, but
later used a towline pulled by boys or horses. An accident ended
his life in 1899 when he consented to make a flight in unfavorable
weather, in order not to disappoint friends who had come to see
him fly.
Of greater importance was the work of Octave Chanute, an

American civil engineer who for several years had studied the prob
lem of flight and had written a critical history of aviation experi
ments. Too old to do any flying himself, he hired several younger
assistants, among them A. M. Herring, who had previously built
and flown a Lilienthal glider. In the summer of 1896 they estab
lished a glider camp in the dune region on the southern shore of
Lake Michigan. First they tried the Lilienthal glider, which they
found dangerous and difficult to handle. The purpose of Chanute's
work was to discover better means of control than that of shifting
the weight of the pilot, and also to obtain some measure of auto
matic stability. So he designed a multiplane with wings that could
swerve fore and aft to adjust the center of lift; it underwent gradual
modifications until quite satisfactory flights could be made. Finally
Chanute designed a biplane which later became famous as the
"Chanute type," and the ancestor of the first successful powered
airplanes. The two cambered lifting surfaces were straight from tip
to tip, and trussed together by a girder of vertical struts and diag
onal wires, known as a "Pratt truss." This construction was simple
and light but very strong; weighing only 23 pounds, it carried 178
pounds. The tail was flexibly attached, an invention of Herring's
which improved longitudinal stability; otherwise control was ob
tained by throwing the legs left and right, but much less strenuous
motions were required than with the Lilienthal glider. More than
a thousand flights were made without the slightest accident during
1896 and 1897. The glider was manageable in winds up to 31 m.p.h.,
and sometimes was lifted higher than the starting point. Most of
the flights were made straight downhill, but Herring also learned
to make a turn and glide lengthwise along the hill, taking advantage
of the slope wind. He reported a flight of 927 feet in 48 seconds
made in this manner. Among the others who started gliding during
Lilienthal's lifetime was a group of young men at Schenectady, New
York, who in 1894 formed the first glider club in the world, the
"Mohawk Aerial Navigation Company," under the leadership of
Charles P. Steinmetz.
Newspaper notices of Lilienthal's death inspired Orville and
Wilbur Wright to study the problem of flight. They decided that
the method of balancing a glider by shifting the weight of the pilot,
as practiced by Lilienthal and Chanute, was not effective and not
the method used by birds. Taking the Chanute biplane as model
for their designs, they made several important changes. To reduce
drag the pilot was placed prone on the lower surface, the tail was
discarded and a front elevator used instead; but most important
of all, they invented a method of warping the wings for lateral

balance. At Kitty Hawk, North Carolina, the Wright brothers made
their first tests in the fall of 1900. Only a few flights were made,
just enough to prove the soundness of their theories and the effec
tiveness of the control mechanism. The following year, with a sec
ond glider, they made a number of good flights. Chanute, who was
a visitor in the camp, said that they had done better than anyone
before, but the brothers were disappointed. The performance was
not up to expectations, and to secure reliable aerodynamic data for
future designs they built a small wind tunnel in which they carried
on tests during the winter.
The 1902 glider, with a span of 32 feet and weighing 116 pounds,
was a larger machine than anyone had dared try before. It was
provided with a fixed vertical tail in addition to the elevator and
wing warping device. During September and October 1902 the
Wrights made nearly a thousand flights with this glider, improving
it according to experience obtained in the air. First they practiced
with the elevator alone, then the warping wires were taken into use.
The operation of two different controls at the same time caused
confusion, but was mastered. Then the glider showed a tendency
to side-slip in the turns, and this was remedied by making the ver
tical tail into a movable rudder. The control of three things at once
seemed too complicated, but was simplified by connecting the
rudder to the warping wires, since both were intended to be oper
ated together. They now had effective control of their glider and
could really begin to learn the pilot's art by continuous practice.
The longest flight was 622 feet and lasted 26 seconds. They did not
try to set records because the purpose of their work was research
rather than spectacular performance. On returning to Dayton they
immediately began work on a powered airplane, for which they
had to build their own motor. In September 1903 they went to
Kitty Hawk again and divided their time there between work on
the power "flyer" and practicing with last year's glider. In gliding
they now became experts, making many flights of more than half
a minute's duration, the longest one lasting 43 seconds. Sometimes
they succeeded in hovering in the strong slope wind over one spot,
once as long as 26 seconds. With the first flights on the motor
flyer, December 17, 1903, the gliding days of the Wright brothers
were ended for a long time.
Chanute, who had continued his experiments to obtain auto
matic stability, ended his active work after sending one of his
former assistants, William Avery, with a Chanute glider to the
St. Louis Fair in 1904. Avery was the first to use a winch for launch-
ing; he was towed from level ground up to 70 feet in the air. His
best flights were less than 300 feet from the point where the hook
was released, because the field was too small. On the first Might in
a better location the towrope snapped and Avery fell and sprained
an ankle.
Another American pioneer of motorkss flight was J. J. Montgom
ery of Santa Clara, California. His work had begun in 1883, but not
much was known about it before 1905, when he gave exhibitions,
launching gliders from hot-air balloons at a height of 4000 feet.
The Montgomery gliders had tandem wings which could be warped
for steering and balance. The flights lasted up to 20 minutes and
included spectacular maneuvering. These exhibitions ended when
one of the operators, a professional parachute jumper, crashed and
was killed due to the breaking of a stay wire.
The work of the Wright brothers was followed with great interest
in France, where Ferdinand Ferber had been working with gliders
since 1898. Ferber, Ernest Archdeacon, Gabriel Voisin and others
were spurred on by Chanute's reports of the American achievements,
which they tried to imitate with indifferent results. The only fact
worthy of notice is that they tried the first auto tow, in 1906.
With the development of powered airplanes, motorless flight was
all but forgotten. Gliders were built, mostly alter the designs of
Chanute and Lilienthal, but no advance was made over the achieve
ments of these pioneers, and the interest in gliding as a sport soon
petered out. Valuable work was done by Jose \Veiss in England and
Igo Etrich in Austria, who used gliders for the study of automatic
stability, but this was mainly for the benefit of powerplane design.
Orville Wright returned to motorless flight for a short time, but
only for the purpose of testing a new stabilizing device intended
for airplanes. In October 1911, at his old flying ground at Kitty
Hawk, he soared 9 minutes, 45 seconds, setting a record which was
not surpassed for ten years. This created a brief revival of interest
in gliding, and some of the leaders in the modern soaring move
ment in America made their first flights in home-made hang gliders
at that time.
In Germany Lilienthal's heritage was taken up in 1909 by a group
of schoolboys in Darmstadt. Like many other youngsters, they used
bedsheets and broomsticks to build primitive gliders, but unlike
others, this group held together until broken up by the World War.
As the boys grew older and some of them became students at the
Technical Institute, their theoretical understanding increased and
their designs improved. In 1912 during summer vacation they dis
covered a wonderful site for motorless flying, on the Wasserkuppe
in the Rhon mountains. Here Hans Gutermuth made a flight of

2700 feet, lasting i minute, 52 seconds.
The World War interrupted all aviation activity other than
military. The airplane was perfected as a weapon, but adaptation
to military needs made it unsuited for the purposes of peace: it was
dangerous for sport and uneconomical for commerce. After the war
Oscar Ursinus, editor of the German magazine Flugsport, started a
campaign for the development of civil airplanes based on sound
aerodynamic design rather than on the brute force of excessive
engine power. Remembering the Darmstadt schoolboys, he proposed
THE FIRST SOARING FLIGHT

Orville Wright at Kitty Hawk in 1911
u' n Brothers
a gliding and soaring meet in the Rhon mountains, for the purpose
of scientific research and healthful sport. Added interest in motorless flight resulted from the restrictions imposed by the Allies upon
powered airplanes in Germany.
The first Rhon competition took place from July to September
1920. Among those who gathered on the Wasserkuppe were engi
neers and scientists as well as amateurs and former war pilots eager
to find a way of satisfying their desire to fly. The combination of
science and sport has characterized the movement ever since. Some
of the gliders were modeled after motorplanes, others were hang
gliders of the Lilienthal and Chanute types. The results were only
fair, until towards the end of the meet Wolfgang Klemperer arrived
with his cantilever monoplane glider "Schwarzer Teufel," designed
by him and built at the Institute of Technology in Aachen. Klem
perer used, for the first time, the shock-cord method of launching,
and surpassed all competitors by remaining in the air for 2 minutes,
23 seconds, covering more than a mile.
Longer flights were made the following year, when Orville
Wright's record was beaten by Klemperer with a flight of 13 min
utes; this was surpassed by Arthur Martens, who flew for 151^
minutes on the "Vampyr," and later Friedrich Hart, whose gliding
experiments started before the war, raised the record to 2114 min
utes. But it was in the third Rhon meeting, in 1922, that things
really began to happen. Downhill coasting was now child's play, and
the experienced pilots set out to master the art of soaring.
Klemperer's "Schwarzer Teufel" had shown the right direction in
WOLFGANG KLEMPERER LANDING His "BLUE MOUSE" IN 1921

design, and the "Vampyr," designed and built by professors and
students at the Technical Institute in Hannover, was the first true
sailplane. On it Martens made the first motorless flight of more than
1 hour, on August 10; the next day Henzen remained in the air for
2 hours, and five days later for more than 3 hours.
Until then motorless flight had attracted little attention outside
the small group of devotees, but these achievements demonstrated
its possibilities to the whole world. A successful soaring meet was
held in September the same year at Itford Hill in England, and
during 1923 there were meets at Biskra, Algeria, at Vauville, France,
and also on the Crimean peninsula in Russia. The French pilots
were, for awhile, serious contenders for first honors: in January
1923 Thoret soared a powerplane with the engine shut off for 7
hours, and this was surpassed first by Maneyrol, then by Barbot,
who held the duration of over 8i/2 hours at the end of the year.
The greatest height1750 feetwas also reached by a Frenchman,
Deschamps, at Biskra. In Germany a new soaring center was estab

lished at Rossitten in East Prussia, where sand dunes along the
Baltic coast provided favorable conditions. At Rhon permanent
buildings began to appear and gliding schools were opened. During
the next few years records were constantly bettered, until in 1925
Ferdinand Schulz made the first motorless flight of 12 hours' dura
tion while the distance record had been increased to 151/4 miles.
However, nothing new had been learned since 1922. A skilled pilot
could soar above a hillside as long as his strength and the slope
]\7olfgaug Klemperer
THE "VAMPYR" FLYING AT HANNOVER, GERMANY, IN 1921

(Now in Deutches Museum, Munich)
wind lasted, but this was not enough to keep the interest alive. The
question was "What next?"
The 1926 Rhon meeting indicated the answer. First Schulz put
an end to the notion that soaring was a fair-weather sport, by flying
in hail and rain. Then Max Kegel was sucked up by a thunder
storm to a greater height than any motorless plane had reached
before, and he more than doubled the distance record from the
previous year. At the same meeting Johannes Nehring in the
Darmstadt sailplane made a goal flight to the Milseburg and back:
it was much shorter than the record, but of great importance

because it was the result of close study of the topographical condi
tions and air currents. Later, under the supervision of Professor
Georgii, Nehring made a number of research flights in very light
winds, and during the following year in upcurrents caused by
houses and trees. These efforts to explore new possibilities for
soaring flight were promoted by the research institute of the RhonRossitten Gesellschaft of which Georgii was the director.
Another important advance was made in 1928, with the beginning
of cloud soaring. The fact that upwinds exist under cumulus clouds
had been known to meteorologists for some time, and Nehring,
flying a light motorplane, had investigated these upwinds. Robert
Kronfeld, a young Austrian, was the first to make use of this knowl
edge when he made contact with a cloud over Wasserkuppe, was
lifted 1400 feet, flew with it to Himmeldankberg, where he hung
in the slope wind until he could make contact with other clouds
and by their help return to the starting point. In the following
year cloud soaring became common practice, and the records for
distance and height mounted rapidly. Kronfeld now made his sec
ond great contribution to the technique of motorless flight by
demonstrating storm-front soaring. Kegel had been carried aloft
accidentally by a thundercloud, but Kronfeld set out deliberately to
explore this possibility for soaring. On July 20, 1929, he started from
the Wasserkuppe in the face of a thunderstorm and established a
distance record of 85.5 miles and an altitude record of 7525 feet.
The widespread interest in motorless flight awakened by the first
sensational demonstrations in 1922 did not last long. Outside of
Germany the movement came almost to a standstill, and even in
Germany it had some lean years, but the great progress since 1926
brought new life. By 1930 soaring societies were active in many
countries, and in that year the ISTUS was organizedthe Inter
national Association for the Study of Motorless Flight.
In the United States, the land of Lilienthal's foremost disciples,
Chanute and the Wright brothers, motorless flight was introduced
again in 1928 by a group of German pilots. They brought with
them the first modern soaring plane seen in this country, the
"Darmstadt," and with it Peter Hesselbach made a flight of over
4 hours on Cape Cod. A glider camp was organized and a wave
of enthusiasm for motorless flight swept the country. Unfortunately
much of the fervent gliding activity which followed was haphazard
and ill-advised, but valuable pioneering work was also done. Even
in this early period Americans were nor. mere imitators of the Ger
mans, but worked out their own methods, At the University of
10
Michigan Professor R. E. Franklin developed the Franklin utility

glider and introduced auto-tow training and launching. Franklin
also practiced airplane tow, first demonstrated by Espenlaub at the
Wasserkuppe in 1926, and in 1930 Frank Hawks was towed across
the continent in a Franklin glider (now in Smithsonian Museum,
Washington, D. C.). Another spectacular feat of that year was Ralph
Barnaby's gliding flight from the airship "Los Angeles."
Outstanding among American soaring pioneers was Hawley
Bowlus. He had built his first hang glider in 1911, and now began
designing and building modern sailplanes. In October 1929 Bowlus
made the first motorless flight of over an hour in an American-built
craft, at Point Loma, California; by February 1930 he had raised
the American record to over 9 hours, and on April 29-30 his assistant
Jack Barstow set an unofficial world record of over 15 hours. An
American distance record of 15.7 miles was established in 1929 by
Wolfgang Klemperer, famous from the first Rhon meetings. He
had come to America as an engineer for the Goodyear Zeppelin
Company, and founded a soaring group at Akron, Ohio.
The first soaring meet in America was held in 1930 at Elmira,
New York, a site which first was explored by Klemperer and Jack
O'Meara. Here, on October 4, Wolf Hirth made the first long
thermal flight, 54 miles without the aid of slope winds or clouds.
With this flight began the systematic study of thermal currents and
the development of thermal soaring technique. Results were evi
dent already in the twelfth Rhon meeting, 1931, when Hirth,
Groenhoff and Kronfeld made flights of over 100 kilometers (62
miles) by this method. Flights were also made in the thermal upcurrents created by great citiesover Berlin, Munich and London
and by Jack O'Meara over New York.
The mastery of thermal soaring liberated motorless flight from
its dependence upon mountains and slopes or chance clouds and
storm fronts. Starting from level country became possible by the
use of new launching methods: airplane tow, first practiced for
exhibition purposes, was perfected as a starting method by Peter
Riedel and Giinther Groenhoff in Germany, auto tow developed in
America, was later brought to Europe, and winch launching was
introduced by Kronfeld.
Airplane tow to 6000 feet made possible Kronfeld's glide across
the English Channel in 1931, which stimulated interest in motorless
flight and brought new life to the movement in Great Britain. The
longest flight of the year, Groenhoff's i7o-mile storm-front flight
from Munich to Kaaden in Czechoslovakia, also started with air-
11
plane launching from level ground. It was unsurpassed for three

years, but was not an official record because this method of launch
ing was not then recognized. Before the end of the year the official
duration record, which had been standing at a little over 14 hours
since 1927, was beaten by Lt. William Cocke who soared for almost
22 hours over Honolulu on December 17-18. Cocke's record was
at the time considered unbreakable; still it was broken in 1933 by
Kurt Schmidt with a flight of 36 hours, 35 minutes.
In the course of a scientific soaring expedition to Brazil, the
altitude record which had been standing since 1929 was beaten by
Fred T. Loomis
WARREN E. EATON
Heini Dittmar in February 1934 when he soared to more than

14,000 feet through three layers of towering cumulus clouds near
Rio de Janeiro. On the same occasion Hanna Reitsch set a record
for women of 7040 feet. She already held the duration record of
10 hours, and shortly afterwards gained the distance record with a
flight of 160 kilometers.
In the United States The Soaring Society of America was formed
in 1932 and under the leadership of Warren Eaton a sound and
well-organized soaring movement replaced the early over-enthu
siasm. America offers excellent natural advantages for motorless
flight, good conditions for thermal soaring and long mountain ridges
12
for slope soaring. Soon American pilots acquired sufficient skill to

utilize these advantages. In 1932 Jack O'Meara set an American
record of 66 miles; in 1933 Richard du Pont soared 122 miles along
the Blue Ridge, and in 1934 he surpassed the world record with a
cloud flight of 158 miles. Lewin Barringer's flight from Ellenville,
New York, to Piketown, Pennsylvania, in April the following year
was 3 miles shorter, but notable as the longest slope-wind flight
ever made.
In Germany the 1934 Rhon meeting brought a surprise: Groenhoff's three year, unofficial distance record was surpassed four times
in two days, Heini Dittmar retaining it with a flight of 234 miles.
All these flights were completed in about 5 hours and were the
result of a combination of thermal upcurrents and great horizontal
wind velocity. Previously thermals had not been expected on windy
days. Now it was discovered that "wind thermals" offer the best
opportunity for long distance soaring.
The 1935 Rhon meet was notable for the great number of long
flights; 209 flights covering more than 100 kilometers. This was the
result of the government program, favoring the training of a large
number of good soaring pilots, rather than a few star performers.
Four pilots, Oeltschner, Brautigam, Heinemann and Steinhoff, es
tablished a new distance record of 313 miles, landing at the airport
of Brno, Czechoslovakia. On the way back to the Wasserkuppe by
airplane tow, Oeltschner was killed in a crash, and to honor their
friend the three others requested that the record be listed in his
name only.
Motorless flight had now been mastered to the extent that long
distances could be covered in motorless planes if the pilot would go
wherever favorable currents might bring him. The next step was
to learn to reach any point he might choose. Goal flights had been
practiced since the first Rhon meetings, but in 1935 for the first
time such flights were made over long distances; in this year Peter
Riedel made a 165-mile goal flight from Berlin to Hamburg, later
Kraft flew 208 miles from Hornberg to Cologne, and during the
Olympic games in 1936 the Hungarian, Ludwig Rotter, made a
goal flight from Berlin to Kiel.
A further development from the goal flight was the sailplane
tour over a predetermined course, which had to be flown within
a specified time and with scheduled stops, regardless of terrain
and weather conditions. The first tour, covering 432 miles over
the route DarmstadtWiiizburgMunchenAugsburgStuttgartMannheim Darmstadt, was completed in 1936 by four pilots. An
other challenge to the skill of soaring pilots was the crossing of the
13
Alps, and this was achieved by six sailplanes, including a two-seater,

during the ISTUS meet at Salzburg early in 1937.
Long goal flights soon became a matter of routine. For example,
one day during the 1937 Rhon contest twenty pilots listed Niirnberg, almost 100 miles distant, as their goal, and nineteen made it.
In 1938 "distance with return to starting point" was added to the
record list, the record holder being Bernhard Flinch with a flight
from Bremen to Liibeck and back, 191 miles.
The 1938 Rhon meet was remarkable for a great number of high
altitude flights. The record was raised to 21,398 feet by Walter
Drechsel, and there were no less than forty flights to over 13,000
feet. Many of these were made by flying blind through thunder
clouds, a feat which previously had been considered extremely reck
less. Before the end of the year the record was raised to 22,560 feet
by Erwin Ziller.
In 1937 records for two-seaters were first officially recognized, and
soon approached those for single-seaters, in the case of duration
even surpassing the single-seater record when in December 1938
A. Bodecker and K. H. Zander remained in the air for 50 hours,
26 minutes.
The motorless flight movement outside Germany had progressed
so far that by 1937 tner countries could offer serious competition
to Germany's best sailplane pilots. This was demonstrated in the
great International Soaring Contest at the Wasserkuppe in July,
where seven nations took part. Germany won the contest on points,
but Poland, Switzerland, Austria and England took prizes although
their pilots did not have the advantage the Germans had of being
thoroughly familiar with the terrain.
Reports came from Russia of great distance flights. In May 1937
Victor Rastorgueff, flying eastward from Moscow, covered first 335
miles, then 374 miles, and finally on May 27, 405 miles. On the
same day V. M. Ilchenko with V. Emerik as passenger established a
two-seater record of 253 miles, and this was raised to almost 400
miles by I. Kartasheff and P. Savtzov in July 1938.
On July 6, 1939, a new world's single-seater distance record of
465 miles was made by a woman, O. Klepikova.
The number of "Silver C's" awarded indicates the progress of
motorless flight in the various countries. In December 1938 Ger
many, far in the lead, had 816; Poland had 159; Great Britain 50;
France 29; Switzerland 19; and the United States 17.
Great Britain twice held the official duration record for twoseaters in 1937 and 1938. A Polish girl, Wanda Modlibowska,
established a duration record for women of over 24 hours in 1937.
14
National soaring contests are now held in many countries, the

Elmira meet in the United States having been held annually since
1930. The official American distance record of 212 miles was flown
by Lewin B. Barringer on April 19, 1938, from Wichita Falls,
Texas, to Tulsa, Oklahoma. It was the first long goal flight in
America and was made from winch launching over level country.
1 In June 1939 Woodbridge P. Brown established a new American
distance and goal record of 263 miles from Wichita Falls, Texas,
to Wichita, Kansas. This was surpassed on July 13, 1940, by John
Robinson who flew 290 miles from Elmira to Mineral, Virginia.
The single-seater altitude record was raised to 17,264 feet by
Robert Stanley on July 4, 1939. On August 12, 1940, Lewin Barringer bettered his national two-place record by soaring to a world
record mark of 14,960 feet over Sun Valley, Idaho, reaching an alti
tude of over 21,000 feet above sea level.
With the erection of permanent buildings at government expense
at the Warren E. Eaton site on Harris Hill at Elmira, and the train
ing there of Army Air Corps pilots, gliding and soaring in the
United States has begun to come of age. In addition to the
Annual National Contest at Elmira there are regional meets in Cali
fornia, Michigan, Texas and New Jersey. With the government
actively sponsoring the training of college students as airplane
pilots, it is likely that this help will be extended to the motorless
field which will furnish the needed impetus to have the sport
increase until thousands of young men and women can take advan
tage of the wonderful soaring conditions that exist all over the
United States.
CHAPTER II
AERODYNAMICS
By Paul Schweizer
Assisted by Milton Stoughton and
Ernest Schweizer
GLIDING FLIGHT
W,HILE THE AERODYNAMICS of soaring flight may at first
glance appear to be somewhat mysterious and complex, in reality

the whole thing resolves in the simple "glide." All the art and
science of gliding and soaring is built up around this simple
phenomenon.
A powerless aircraft is said to be "gliding" when it slides along
and down through still air in the same manner that a sled slides
down the snowy slope of a hill. While it may at times be sliding
on a steeper angle than at others, it is always coming downhill in
this air. Having no power of its own to propel it, there is no other
way to maintain forward motion except by letting its own weight
pull it ahead just as a sled or a cart moves down a hill. The only
difference is that in this case the wings are the "wheels" or "run
ners," and the air is the "hill" on which it slides. Just how fast it
will glide downhill, and how steeply, depends upon the design of
the glider and on how the pilot controls it.
The very efficient high performance gliders can glide a long way
without losing much height. In fact, some can "coast" 30 miles in
still air for every mile of altitude lost, without the help of anyrising air currents such as make it possible to keep a glider up for
hours at a time. In Fig. i is a comparison of the glides of various
types of gliders.
'Tr '*"'"'
h PWMAr
10 mat
FIG. i. AVERAGE GLIDES FOR DIFFERENT TYPES OF GLIDERS

IN STILL AIR
Angles shown are not true gliding angles
16
The altitude that a certain glider will lose in a given time is

called the "sinking speed" and depends upon the design of the
glider and upon how fast the pilot flies it, as the gliding ratio, as
it is called, changes with the airspeed. There is always a certain
airspeed at which the glider will "sink" the slowest, and another
speed, slightly faster, that carries the glider the greatest distance
when starting from a given altitude. The slowest rate of descent
is called the "minimum sinking speed" and is usually given in feet
per second. The sinking speed of a modern high performance sail
plane is about 2 feet per second, while as little as i foot is possible.
Since it is desirable to have a good forward speed and "glide" in
order to make long distance flights, gliders are not always designed
to have the lowest possible sinking speed. This apparent paradox
will be explained in detail later on in the text.
NORMAL GLIDING FLIOH,

IN STILL AIR
RATE OF DESCENT* JINKING SPEED
SOAPING F:.IGHT /,v UPCURRENT

RATE or CLIMB vu -vs = vc
UPCURRENT GREATER THAN
SINKING
LEVEL FLIGHT IN UPCURRENT

RATE OfCUMB=0
UPCURRENT = SINKING SPEED
GLIDING IN QOWfJCLiRRENT
RA TE OF DESCENT l + Vk V0
DOWNCUXRENT ADAS
TO SINKING SPEED
FIG. 2. GLIDING AND SOARING FLIGHT

Soaring flight differs from gliding flight in that the machine flies
without losing altitude, and even climbs higher. This is possible
because the pilot is flying in a part of the air that is rising bodily,
carrying the glider in it, just as a leaf or paper is carried up by the
wind. The soaring glider is still sliding downhill in this "body" of
air, but not so fast as the whole is rising. So the basic principle of
soaring is to get in an upcurrent that is rising more quickly than
the glider is going down. The difference between these two will be
the resulting climb. In Fig. 2 the various phases of gliding and soar-
AERODYNAMICS
17
ing flight are shown. The art and skill of soaring come in finding
and making the most of these rising air currents to gain altitude
and fly distances. From this it is evident that the success of a flight
depends not only on the performance of the glider but also upon
the skill and knowledge of the pilot.
A knowledge of simple and basic aerodynamic principles and
formulas will help a great deal to understand the principles of
gliding and soaring flight. As figures and equations "frighten"
many, the following explanations have been made as simple as pos
sible and yet still contain the actual formulas and equations used
in soaring aerodynamics. Of course, for simplification, many of the
intermediate steps have not been shown. The average reader can
take these for granted and the more advanced readers probably
know them.
A glider in flight is supported by the lifting effect of its wings
as it passes through the air. This lift is proportional, to some ex
tent, to (a) the size of the wing, (b) the speed that it moves through
the air, (c) the weight, or density, of the air, and (d) the particular
aerodynamic characteristics of the wing.
It seems logical that the weight or density of the air (the close
ness of the air particles) will affect the lift, as the forces on the wing
are due to the reaction of these particles on the wing. The greater
the density of the air, the more particles can act against the wing
with resulting greater force. So it is evident that the lift must be
directly proportional to the density.
The lift is also directly proportional to the area, as the amount
of wing that can react against the air is proportional to the area.
If we have a wing that is twice as large as a given wing in area, it
will have twice as much lift.
Unlike the two previous factors, the lift is proportional to the
square of the speed, for an increase of speed not only brings a
greater amount of air past the wing but also increases the energy
that the air particles can give to the wing. For example, if the
speed is doubled there is twice as much air passing over the wing
at twice the speed, which means that the lift is two "squared" or
four times as much.
The lift of the wing is very much dependent upon the shape of
the wing, or airfoil section, and also upon the angle of attack, or
angle of inclination of the wing against the wind. For simplification
this variation of the lift with the angle of attack and shape char
acteristics (airfoil section) is called the lift coefficient. This coeffi
cient also eliminates the question of airspeed when discussing these
characteristics.
18
Putting all these facts into a formula and setting them equal to
the lift, we get:
Lift L = K p Cl S V2
where p = density, Cl = lift coefficient, S area, V = speed and
K is a constant to take care of the units of these factors. This lift
equation is the basic flight equation.
The same line of reasoning can be used to prove that the drag
is dependent upon the same factors, and so we get:
Drag = D = K p Cdt S V2
where Cdt is the total drag coefficient for the glider and all the
other factors are the same as in the lift equation.
In Fig. 3 is shown a glider in normal gliding flight. The angle
+HOR
GLIDING ANGLE &
FIG. 3. FORCES ON GLIDER IN FLIGHT

between the line of flight and the horizontal is called the gliding
angle 6 and the angle that the line of flight makes with the refer
ence chord of the wing is called the angle of attack. Now the
lift and drag forces derived above act in gliding flight and are re
acted by the weight of the glider. The lift acts perpendicular to
the line of flight and the drag parallel to the line of flight. The
weight which acts straight down is broken into two parts; W sin 6
which is the part that pulls the glider along its path, and W cos 6
which is the part of the weight that the lift has to support. R is
the resultant air force and it is due to the lift, drag and tail bal
ancing loads, and is equal to W. In a steady glide all these loads
balance so that we can set them equal to each other.
W sin e = D
W cos 6 L
Dividing these by each other we can get them into a very con-
AERODYNAMICS
19
sin 0
D
,.
sin 0 D
\V
r
, form.
terms
-_canceling
venient
cos<9
L
&
Wcos<9~L
D
sin f)
so =z tan 0
From trigonometry tan 0
L
cos 0
Also from previously derived formulas:
D _ K p Cdt S V2 _ Cdt
L "~Kp Cl SV2 ~~~CT
Cdt.
Since -=- = tan
Or from trig, cot 0 =
then tan 0 =
t*
CT
This shows that the angle of glide depends upon the L/D ratio
of the glider and not upon the weight of the glider. It is entirely
a question of aerodynamic efficiency and does not vary with weight.
The L/D ratio is really an efficiency ratio as it shows how much
useful lift we can get for a given amount of drag.
Now, referring to Fig. 4, we can get some relation for the sink/VORMAL GLIDE
(NOT ro SCALE:)
FIG. 4
ing speed of the glider. The velocity of the glider along its glide
path is Vf and the sinking speed, or velocity downward, is Vs.
Vs
By solving this system by trigonometry we get sin 0 = .
However, in the range of gliding angles 0 for gliders the sine is
substantially equal to the tan.
So we can put tan 0 = =j~, but as tan 0 also equals,
VI
Vf
Vs
D
we can set = ^77 or solving for VS =
JLj
VI
L-i
Li/iJ
or putting it in words, the sinking speed is equal to the glider's

speed divided by the gliding ratio.
For an example take the glider in Fig. 4. It is traveling along
its glide path at 60 ft./sec. and its L/D is 20. From the L/D it is
evident that for every 20 feet that the glider moves forward it
loses one foot in altitude. So in i second it will travel 60 feet or
lose 3 feet per second, which is its sinking speed. The same result
can be obtained by using the sinking speed formula that we derived
and divide the speed (60) by the L/D (20) which will give us
3 ft./sec. sink.
20
Sinking speed is the basic formula of soaring and until a few

years ago the lowest possible sink was the goal of every designer
of high performance sailplanes. The plane that had the lowest sink
could rise highest on upcurrents and stay up on weak currents
that would not enable gliders of greater sinking speed to soar.
Today, due to the popularity of cross-country flying, other factors
are also important for cross-country gliders.
In order to see upon what sinking speed depends we will put
the equation into a different form.
Our original form is Vs =
L/D
From above ^ = ~ and L K p Cl S (VS) 2

As the lift is substantially equal to the weight for usual gliding
angles we can substitute the weight for the lift.
___
So W = K p Cl (VS) 2
/wr~~
/w
or solving for Vf = -v/
\ JV p V><1 ij
= A/
\ iJ IV U \^>L
substituting for Vf in Vs =
_
__ _
/W i
/W_KCdt /W
/ i
\SKpCl _Cdt
Cl/Cdt~~CP^ \ K p \ S "Cl 1 - 5 \ S
From this we see, assuming that the other factors stay the same,
that a reduction in weight or an increase in wing area without a
corresponding increase in weight will lower the sinking speed.
This is evident from the term W/S which is the wing loading (the
total weight of the ship divided by the area). Decreasing the weight
of the ship is the most obvious method to decrease the sinking
speed and was about the only approach used by the early glider
pioneers.
Now if the formula is put into another form we will see how the
problem of reducing sinking speed was approached from a different
angle with much improved results. A new term, aspect ratio, now
comes into the discussion. This term is really a slenderness ratio of
the wing and for a rectangular wing it is equal to the span divided
by the chord. However, with curved and tapered wings this ratio
is equal to the span 2 divided by the area, a more general formula
that can be used for any type of wing. From this it is evident that
the higher the ratio the more slender the wing.
Changing the form of the aspect ratio equation we get
T) o
"R2
A.R. =-rArea S = .=-where B is the span.

l
A.R.
Area
AERODYNAMICS
Substituting this in formula No. i for sinking speed we get:

v
KCdt lW~
KCd
Cl
AR
In this form, again assuming that the other factors remain un
changed, we see that an increase in span will decrease the sinking
speed considerably as it is to the first power while the wing loading
of the previous form was to the half power. The term W/B- is called
the span loading and is equal to the weight divided by the
span squared. A low span loading is a good indication of low
sinking speed and, as will be explained later, it is more important
than aspect ratio for minimum sinking speed, although one is
dependent upon the other.
The early designers had carried lightness to the extreme, follow
ing along the first line of reasoning, and any further development
along this line did not yield much improvement but seriously
endangered the strength of the gliders. Following the second line
of reasoning increase of span resulted in much improved sinking
speeds. This also brought improvements in gliding angles and speed
characteristics which spurred development.
The previous discussion has taken place under the assumption
that the drag and lift coefficients stayed the same while the other
factors varied. Actually these coefficients vary greatly with design,
and also with span and aspect ratio. In fact improving the lift and
drag characteristics is highly important in lowering the sinking
speed and general performance. The following discussion will show
what factors determine these lift and drag coefficients.
The lift and drag coefficients as they appear in the previous two
formulas are for the complete ship. As the other parts of the glider
contribute very little, if any, to the lift, the lift coefficient is just
for the wing. The drag coefficient is composed of the drag of the
wing and also the other various drags of the glider. The lift coef
ficient will be discussed first.
As mentioned before the lift is very much dependent upon the
wing shape and airfoil section. The first airplanes had flat surfaces
for wings which derived lift from the air stream hitting their
inclined surface, much as a kite flies. However, it was soon dis
covered that by curving the wing and giving it two surfaces, top
and bottom, the lift could be greatly improved and in present-day
airfoils most of the lift is due to the action of the top surface of
the wing and only a small part to the effect of the air stream hitting
the bottom surface.
By varying the shape of the airfoil section, the characteristics
22
can be radically changed. There are thousands of different airfoil

shapes available, each with its special features. The designer chooses
the one that suits his purpose best or else designs one of his own.
In Fig. 5 a set of curves is plotted in the conventional manner for
an airfoil. The lift and drag coefficients, L/D and center of pres
sure are plotted against the angle of attack or speed as each angle
is proportional to a different speed. As the angle of attack increases
-8
-8-4
048/2 16 20 24 28 32
Angle of attack, a (degrees)
FIG. 5
the speed decreases and the lift coefficient increases. This is evident
from the formula for lift, L = K p Cl S V-. For a given plane in
level flight the only variables are lift and speed, as the wing area,
density and constant are fixed at any altitude. So if one of these
factors is increased the other must decrease in order to balance.
As the lift coefficient increases with angle of attack there is a
point where the curve suddenly changes direction and the lift goes
down. This is known as the stall. Here the airflow around the wing
AERODYNAMICS
23
suddenly changes form and the lift is very much decreased. In

flight this sudden loss of lift causes the ship to drop. Some airfoils
will cause the ship to drop relatively slowly, or "mush," while
others will lose most of their lift and drop the ship much faster.
As a rule when stalled the nose of the ship drops and with the
loss of some altitude it will pick up speed and lift again. The
main danger of the stall lies in the fact that most gliders and
airplanes will start to autorotate or "spin" in the stalled position
if some gust or unbalanced weight or force should cause the plane
to start turning. So it is important that the plane is in rig; that
is, so balanced that it wants to keep in a level position and not
turn off to one side or drop a wing. However, by special design
and/or by the use of special devices, a plane can be prevented
from stalling and hence eliminate the spin.
In using the airfoil curves in Fig. 5, it is important that the
scale effect be considered. As these curves are determined experi
mentally in the wind tunnel with small models under different
speeds, pressures and other conditions, it is important that the
results be corrected to the conditions of the glider. In order to
facilitate this, a factor called the Reynold's Number (R.N.) is
used to correct for this, and it is dependent upon the speed, size
and density of the air. Knowing the size of the glider and condi
tions of operation, the R.N. can be determined from the formula
R.X. =: 6350 v 1 where v is the speed in ft./sec., 1 is the wing
chord in feet, and 6350 is the constant for standard air. Knowing
the R.N. of the glider, one can then get data at this number or
correct to this condition.
Of all the wing characteristics, the maximum lift is about the
most susceptible to scale effect. As a rule one never gets the maxi
mum lift as given in the usual Cl curves. This means that the
stalling and landing speed are higher than would be expected. The
shape of the peak of the lift curve and the type of stall are also
closely tied in with scale effect and vary considerably for large
variations in R.N. Other factors such as drag, moment coefficient,
center of pressure, etc., do not show so much change. It might be
mentioned here that the maximum lift and type of stall depend
also upon the wing plan form, surface smoothness, interference and
other factors.
In choosing an airfoil for a glider there are many features that
enter into the choice, such as: the L/D, maximum lift, minimum
drag, moment (twisting) coefficient, allowable depth for spars, scale
effect, ease of construction, etc. Of course the design and purpose
of the glider determine which of these factors are most important.
24
However, the most obvious indication of a good airfoil is its effi

ciency ratio or L/D ratio. The L/D listed in the airfoil reports
is only for the wing and does not include the drag and lift of
the other parts of the ship. These parts add little, if any, to the
lift of the ship but add considerably to the drag. This causes the
L/D ratio to drop considerably.
The airfoil data given are usually for a wing of aspect ratio of 6
and so must be corrected to the aspect ratio of the glider. If it
is over 6 then there will be a decrease in the drag due to the
decrease in drag with increase in aspect ratio. But this is a problem
in drag and will be explained below.
The total drag coefficient is divided basically into what is called
the "parasite" drag (it does not result from a directly useful func
tion) and the "induced" drag which results directly from the lifting
force on the glider. The parasite drag is made up of the "profile"
drag of the wing and the form drag of the fuselage, tail, and any
other projecting or exposed items that cause a disturbance of the
airflow. This drag also includes the interference between fuselage
and wing, wings and struts, etc. It does not include the effect of
the presence of the fuselage on the flow over the wings, which can
alter the induced drag due to the lift, by changing the distribution
of loading over the wings.
While the profile drag coefficient of the wing is entirely a matter
of the sectional shape of the wing section and secondarily of size
(scale effect) and attitude, induced drag is a matter of the wing
shape in plan and of the load distribution over the wing span.
The drag results directly from deflecting the air downward behind
the wing to provide lift. At slow speeds the air must be deflected
down more sharply than at high speeds, so the induced drag coef
ficient is high at high angles of attack of the wing (low speeds)
and low at the low angles of attack (high speeds). Thus at high
speeds in a powerplane the induced drag is almost negligible.
However, in glider design, especially for those not designed for
high speed cross-country work, the induced drag is never negligible
in the normal operating speed range.
This will be clearer if we study the formula for the induced
drag coefficient. The induced drag coefficient = Cl2/-iAR where Cl
is the lift coefficient of the wing and AR is the aspect ratio of the
wing. This shows how it is most important at high lift coefficients
and that it is decreased with larger aspect ratios. As the lift is sub
stantially fixed the induced drag is lessened by increasing the aspect
ratio. This is one reason for the comparatively large span of sail
planes.
AERODYNAMICS
25
Since the weight of the wings runs up rapidly with increasing

aspect ratios, and the maneuverability is reduced, a compromise is
usually established between these conflicting factors. The best de
sign is the one resulting from the wisest choice of compromises: but
perfection is unattainable as requirements and results are still a
matter of opinion.
In general, the effect of high aspect ratio is to reduce the induced
drag. The elliptical wing shape is the theoretical ideal, but it can
be approximated closely by wings with a straight center section
and tapered tips. The straight wing and the too sharply tapered
wing are not so efficient as the elliptical or properly tapered wing.
A poorly located or improperly faired fuselage can alter the dis
tribution of air loading over the wings and have the effect of
increasing the induced drag. It is also possible for the fuselage
interference to decrease the induced drag by offsetting the effect
of a too sharply tapered wing.
The fuselage and tail surfaces are the remaining important
parts causing the parasite drag of a glider. A well shaped fuselage
may have scarcely more drag than the tail surfaces. Its drag consists
of a combination of the effect of the surface area (skin friction)
and the form or shape of the fuselage. If the shape is very good,
the total drag approaches that of pure skin friction. This latter
value varies with the scale effect and the quality of surface finish.
In this respect it behaves similarly to the profile drag of the wing.
At high speeds especially, the smoothness of a surface is an im
portant factor in performance. This is true particularly on aircraft
of high speed design where small items can loom large in power
losses. In gliders, of course, although it is not so important, these
losses appear as reduced gliding range and increased sinking speed.
With respect to the shape that produces the least drag in a
streamlined body or strut, it is interesting to note that it is the
forward portion that is the most sensitive to variations in shape
or to small interferences which cause local turbulence. For this
reason great care should always be exercised to obtain the least
possible disturbance of the natural lines of a fuselage when design
ing windshields or closed cabins. The intersection between the
wing and the fuselage is also of primary importance, since improper
arrangement here will also cause unnecessary drag and interfer
ence with the wing lift distribution. Fillets between the wing and
fuselage intersection are usually necessary for low wing positions
to avoid drag and stalling difficulties. On mid-wing and higher
locations the problem is not so difficult. In many cases the best
solution is no fillets at all.
26
A well shaped fuselage section is more or less rounded so that

there are no very sharp corners to cause drag when the air is flow
ing at an angle to it, and whose longitudinal shape is approxi
mately an ellipse forward of the maximum section and a parabola
aft to the tail. The shape of a symmetrical airfoil section expanded
to the desired depth also makes a good fuselage form. In general,
the exact shape of the basic form is far less important than the
nature of the disturbances and interferences caused by the addition
of wires, cockpit enclosures, etc., since these usually have a powerful
effect on the character of the airflow over the combination of parts.
This principle also applies to wings where it is more important
to hold the contour of the nose of the airfoil so that there is no
break in the true curve over the spar, where the fabric is usually
attached, and to provide a highly finished surface, than it is to
use the most efficient airfoil section. The theoretical gains are
small in comparison to the gains that can be obtained with a little
care and high grade workmanship. The avoidance of round struts
or wires, very oblique struts, Venturis, and other "drag producers"
will help to increase the performance.
STABILITY
A glider can fly steadily on a straight path because of its "bal
ance" and "stability." Both conditions are obtained by means of
the tail surfaces or by special design of the wing to get the same
effect on a tailless glider. Since the air forces on the wing are
not the same at all angles of attack but shift fore and aft, and the
center of gravity is fixed, the glider is balanced at any desired
angle of attack by means of the tail surface "elevator." By tilting
the elevator, the load on the tail is varied to obtain any desired
angle of attack, which in turn fixes the airspeed. Thus, at low
angles of attack (low lift), the airspeed becomes higher to sustain
the weight and at high angles of attack (high lift) the airspeed
is lower. A rigidly fixed tail surface will balance the craft at only
one angle of attack and airspeed providing the weight is not changed
nor the location of the center of gravity altered. This characteristic
can be noted in flying models.
However, a glider might balance but still be unable to make a
steady flight for the lack of sufficient stability. After being dis
turbed in flight the ship should right itself and fly steadily without
too many oscillations. The fixed portion of the horizontal tail
surfaces is called the stabilizer and its purpose is to keep the glider
from diving or stalling. It must have sufficient area and the dis
tance that it is placed back from the center of gravity with respect
AERODYNAMICS
27
to the wing is also of importance in obtaining longitudinal stability

(stability up or down along its flight path).
The stability necessary to keep the aircraft from turning off its
straightaway course is furnished by the fixed part of the vertical
tail surfaces called the fin. This is the same as the weathercock
effect on the familiar weathervane. The rudder on the vertical
surface acts the same as the fin whenever it is held stationary by the
rudder control.
To keep the glider from rolling off sidewise into a "sideslip,"
dihedral angle is supplied to the wings. The wing has dihedral
angle when the tips are higher than the center of the wing. This
may be a very small and almost negligible angle in some cases,
but the present trend is toward more dihedral to give better sta
bility in circling.
In general the more stable an aircraft is, the slower and more
difficult it is to maneuver. Consequently experience has taught what
minimum amount of stability is satisfactory for various types of
airplanes and gliders, without introducing other disadvantages.
Training gliders emphasize stability while acrobatic sailplanes fea
ture maneuverability. Special care is necessary to avoid obtaining
"spiral instability" by having too much directional stability com
pared to lateral stability. This results in a condition where the
plane tends to tighten itself into a sharp spiral flight unless it is
held out of it by the aileron control, when making a normal turn.
The opposite condition of too much lateral stability relative to
directional stability results in a sort of wallowing motion commonly
known as the "Dutch Roll." This combination is not so common
as the other, however. If a rudder and fin is made large enough
to provide the needed directional control, there will usually be
plenty of directional stability. Sufficient dihedral angle is then
provided on the wing to avoid spiral instability.
Lateral stability also depends upon the characteristics of the
wing design besides the dihedral angle, particularly around the
stall. On ordinary tapered wings the tips stall first and as the stall
moves in along the span the ailerons become ineffective in this
stalled region. Some dissimilarity of the wing or a gust will cause
one wing to drop or "fall off," and as the ailerons are ineffective,
the glider may drop into a spiral dive or spin.
This can be avoided by using "aerodynamic" twist, which is the
change of section along the span so that the tips will stall later
than the center part of the wing. This can be done more easily,
but not quite so efficiently, by using "geometric twist," which is
the actual twisting of the wing so that the tip is working at a
28
smaller angle of attack than the center and hence stalls later. The
second method is the most practical and is called "washing out"
the wing. In training gliders it is desirable to have the wings
"washed out" and if it is not built into the wing some provision
is usually made to adjust it.
The twisting of the wing definitely helps aileron control over
the whole flight range as the ailerons are always operating at lower
lift. Twist may also improve performance slightly by decreasing
the induced drag. This is due to the possibility of getting elliptical
lift distribution from a rectangular wing for one angle of attack,
by using the proper twist. The general effect is to reduce the lift
at the tips which reduces the induced drag.
PERFORMANCE
Fig. 6 shows the general performance curves at sea level of a
clean intermediate class sailplane. The curves plotted are the L/D
and the sinking speed against the airspeed. These curves are typi
cal, with a rapid increase of L/D and sinking speeds at the higher
values of forward speeds. It is interesting to note on the sinking
speed curve that the best forward speed to fly for minimum descent
is approximately 5 m.p.h. above the stall. This figure holds in
general for most gliders and shows the fallacy of flying the ship
near the stall for minimum sinking speed.
FIG. 6
AERODYNAMICS
29
The speed for flattest glide is about 7 m.p.h., in this case, above
the minimum sinking speed. In very efficient sailplanes this spread
would be much greater. The best general "soaring" speed is usually
taken as being half-way between the minimum sinking speed and
the maximum gliding ratio. For this model it is just about 36
m.p.h. The cruising speed is the speed at which the sinking speed
is 6 ft./sec. This has been arbitrarily set at this value and it is just
a means of comparing the high speed qualities of sailplanes. The
better cross-country ships will have a higher speed for the given
6 ft./sec. sinking speed.
The curves in Fig. 6 are for sea-level operation and are sub
stantially correct for normal soaring heights. At higher altitudes
the actual sinking speed and flying speed will be higher, but the
value of the gliding ratio is unchanged, it being a function of the
design of the glider and angle of attack of the wing.
L _ CL
**
SAMPLE PERFORMANCE CALCULATIONS

___CL
CD Total
CD profile + CD parasite + Înduced

i
2\V
/2W
VF.P.S. =
X
Sp CL
\Sp
VCL
o
b"
VF.P.S.
Sinking Speed =
13.9
A.R
\j <j
L/D
s
146
i .21 X 1.25
c*
Tt-i>*.
parasite
W
S
____
467
r
'
146
q r>
J '~
p = .002378
2 X 467
.P.S. = "V 146 X .002378
CD ;
induced
= .01035
TT' (A.R.)
(3)
0)
(2)
CL
CDprof.
Cor
rected
Cind
.088
.00385
5>-9
VCL
VCL
Correction for
Plan Form = 1.05
(4)
(5)
'(6)
CD ind(7)
Sinking
Speed
(4)
VF.P.S.
51- 9
VCL
17-38
82.1
4-73
Ctot.
(2) + (3)
+ CD par
L/D
'(0
.023
(5)
L X 1.05
TT (AR)
(8)
VM.P.H.
6
1.47
.1
3
4
FIG. 7
55-8
30
FLI6HT WITHOUT POWER
This also means that each "indicated" airspeed on the dial of

the instrument corresponds to a certain angle of attack of the wing
and that the glider will always stall at the same point on the dial.
It can be shown also that the best speeds to fly for the minimum
sink and the maximum gliding ratio will be at the same "indi
cated" airspeed on the dial.
The means of determining the performance of gliders and sail
planes is not difficult and the general method will be explained.
In Fig. 7 appear the formulas and sample calculations for the same
ships as in Fig. 6. The formulas used are those that we have already
developed. It is important to note that Cd total equals the sum of
the profile, parasite and induced drag coefficient.
It is evident from this that the calculations for speed are definite
and accurate except at high angles of attack where the question
of stall comes in. This is not important, however, as the per
formance around this speed is not desired.
The big question in performance calculations is the evaluating
of the drag coefficients. The Cd profile must be corrected for the
effects of R.N. This may be appreciable and it is best if one has
data at the R.N. of the glider. In general the surface of the actual
glider wing will not be as smooth as that of the model, necessi
tating an increase in profile drag, to be estimated by the designer
from such data as he can obtain. The induced drag as given by
the general formula is for the ideal elliptical wing and must be
corrected for all other plan forms. The elliptic wing has the mini
mum possible induced drag and all others have an increasing
amount depending upon how far they vary from the true ellipse.
By applying a correction factor we can get an "effective" aspect ratio
which is somewhat lower than the actual of the wing and so results
in greater induced drag. The value of these corrections varies with
plan form. For a rectangular wing it may be from 6-10% for aspect
ratios of 5 to 12. For tapered wings which closely approximate
the ellipse it will be less than half this amount. Data are available
in aerodynamics texts and NACA reports enabling the designer to
make a reasonable estimate.
The parasite drag is the most difficult to determine exactly
because of its complexity. It consists of the drag of the fuselage,
tail surfaces, and bracing, and the effect of interference between
parts. The drag of the tail surfaces may include induced drag also
besides the profile drag, as the lift on the tail surfaces necessary
to balance the ship causes induced drag.
In a sailplane with full cantilever wings and tail surfaces, and
no serious external protuberances such as control horns, fittings,
AERODYNAMICS
31
open cockpits, etc., the problem of parasite drag becomes relatively

easy. It is then just the addition of the fuselage drag and tail
surface drag plus the interference drag between fuselage and wing
and fuselage and tail surfaces. All these can be determined readily.
In a glider of much lower performance the many struts, wires,
open cockpits, exposed fittings, etc., all make the estimate of drag
coefficients more difficult. However, in such ships approximate per
formance figures will suffice. At present very little data are available
on sailplane fuselages, but the drags can be estimated from data
on similar shapes in various references. It is convenient to compare
parasite drags of gliders and other aircraft by expressing it in terms
of flat plate area or drag area. The drag area is the area of a flat
plate which has the same drag as the ship in question. The drag
coefficient of a flat plate is approximately 1.25 per sq. ft. The per
formance curves given as an example are based upon a flat plate
area of 1.21 sq. ft. It is usually convenient to express this in terms
of wing area and so the coefficient obtained by multiplying the
area by the drag coefficient is divided by the wing area. This puts
the coefficients in the same units as the other drag coefficients of the
wing and it can be added directly.
The calculations are worked out for Cl equals .4. The Cd profile
at this Cl is taken from the airfoil curves and corrected for scale
effect if necessary. The Cd induced is determined from the regular
induced drag formula multiplied by a factor of 1.05 to correct for
plan form. The parasite determined above is added to these two,
giving the total drag of the ship. The L/D is determined by divid
ing the lift coefficient by this total drag coefficient. From the
standard flight equation the speed for Cl equals .4 is determined.
Dividing this speed by the L/D will give the sinking speed in
ft./sec. This is worked out for each Cl and the results are plotted
as in Fig. 6.
The high performance of the modern sailplane results mainly
from the reduction of drag. This is accomplished by designing the
fuselage as "clean" and small as possible with all unnecessary
protuberances eliminated, and selecting the most suitable wing sec
tion, plan form, span, aspect ratio, etc. It becomes very important
to reduce the induced drag in order to get high performance. This
is accomplished by using a comparatively large wing span, giving
a low value of "span loading," which is the key to a low induced
drag. This also results in a high aspect ratio, but it is the span
loading rather than the aspect ratio which reduces the drag.
Fig. 8 is the drag curve of the glider whose performance is
given in Fig. 6, with the actual drag in pounds plotted against
32
airspeed. The drag is divided into its usual three parts: induced,
profile and parasite. This shows how the drag is divided and how
it varies with the speed. The induced drag decreases as the speed
goes up because it is dependent upon the lift coefficient, so that at
FIG. 8
high speeds the induced drag is only a very small per cent of the
total drag. The profile and the parasite drag, however, increase
with speed.
It is interesting to note that the performance of a given glider
can be improved at speeds above optimum by increasing the wing
loading. This has the effect of moving the L/D curve in Fig. 6
horizontally to the right. Thus if the glider in the example were
loaded so as to move the L/D curve 5 miles to the right, at 60
m.p.h. the ratio would be increased from 16.2 to 18.2, and the
sinking speed would be reduced from 5.4 f.p.s. to 4.8 f.p.s., or a
reduction of .6 f.p.s. This sort of overloading is useful only on long
fast flights, and cannot be done unless there is a margin of strength
in the structure to carry the overload safely.
CHAPTER III
TYPES OF GLIDERS
By Lewin B. Barringer
IN THE EARLY DEVELOPMENT of motorless, heavier-than-air

aircraft there were a number of different glider types. The first suc
cessful type was the hang glider, so named because in flight the pilot
hung from his armpits or elbows. There were no movable control
surfaces on these primitive types, which could only be balanced
statically by the pilot shifting his weight by moving his legs. The
first successful hang glider built by Lilienthal was a monoplane, but
he later also developed a biplane type. Octave Chanute experi
mented with multi-wing designs including one with five superim
posed planes. His greatest success, however, was with the biplane
type later brought to a higher degree of development by the Wright
brothers who built movable wings and tail surfaces for control and
lay on the lower wing instead of hanging through it.
Another type of hang glider was the tandem built by Montgom
ery in California. This design had two planes mounted one behind
the other with the pilot hanging between them from the framework
by which they were attached. It is interesting to note that a contem
porary Californian has built a biplane hang glider to experience
the thrill of taking off after self launching by running down a hill.
This glider has conventional control surfaces actuated by a small
control stick. Landing is made on a single wheel and tail skid. Its
use is limited to short gliding flights.
Although the most successful early types were biplanes, all
modern gliders are now monoplanes divided into four types: the
primary glider, the secondary or utility glider, the intermediate sail
plane, and the high performance sailplane. The term sailplane is
generally used to denote a glider with performance capable of real
soaring flights such as the requirements of the "Silver C" license.
There are no sharp dividing lines between these types, but their
general characteristics are distinct enough for the accepted classi
fications.
THE PRIMARY GLIDER
The primary glider is the lightest, least expensive and simplest in
construction of the glider types. It has a high monoplane wing and
33
34
FLI6HT WITHOUT POWER
tail surfaces braced by wires or struts and an open truss-braced

fuselage. The pilot's seat is entirely open and unprotected.
The span of the square, untapered wing of the average primary
is about 34 feet with an aspect ratio of 7 to i. Its weight empty is
about 175 pounds. With a wing area of 170 square feet the wing
loading in flight is under 2 pounds per square foot. The gliding
ratio is about 12 to i with a sinking speed of 4 feet per second.
Stalling speed is 20-23 m.p.h.
Philip Ellicott Rarringcr
A WACO PRIMARY GLIDER (1930-1931)

The fuselage construction of the average primary is entirely of
wood with the exception of the metal fittings. Several makes have
been built, however, with fuselages of brazed or welded light steel
tubing which are capable of standing up somewhat better under
prolonged abuse of student training. Wing construction is of wood,
fabric-covered, employing two spars. Drag loads within the wing are
taken by cross-bracing of wire or wood.
The use of the primary glider should be restricted to airport glid
ing and instruction through the "B" license stage of making 360
turns from heights up to 400 feet. Although successful soaring flights
have been made with primaries, they should not be used for this
type of flying as they lack sufficient strength, stability and control
for adequate safety in the turbulent air conditions sometimes found
in slope soaring. There is no protection for the pilot in case of a
crash.
The cost of a new primary glider is about $385. It also can be
purchased in kit form for home assembly at a price around $185. A
lew good used primaries are sometimes available at prices of $100
upwards.
TYPES OF GLIDERS
35
THE SECONDARY GLIDER

The secondary, or utility glider, as the American version is gener
ally called, is more refined than the primary in that it has an en
closed fuselage, a more efficient wing braced by streamline struts
and a pneumatic landing wheel equipped with a brake. This type
is rapidly gaining favor over the primary for training despite its
higher cost. The reasons for this are its superior advantages with
respect to pilot protection, ruggedness, control and performance.
For most clubs the last mentioned consideration is probably the
deciding one as the utility type can be used for both primary train
ing and soaring. It is ideal for slope soaring and can also be used
for thermal soaring, although its slow forward speed and lack of
maneuverability as compared to a sailplane make it rather ineffi
cient for this advanced flying.
The utility usually has a rectangular wing with rounded tips, a
span of about 36 feet and an aspect ratio of 8. Its weight empty is
about 220 pounds which, with a disposable load of 200 pounds for
the pilot, parachute and instruments and a wing area of 180 square
feet, gives a wing loading of from 2.5 to 2.8 pounds per square foot.
The gliding ratio is about 15 to i with a sinking speed of 31/2 feet
per second. Stalling speed is 24-26 m.p.h.
Hans Groenhoff
STEVENS-FRANKLIN UTILITY GLIDER

An interesting conversion of a utility glider is the StevensFranklin. Tapered, gull wings of 48 feet span with the same wing
area as a standard Franklin utility were developed by a group of
engineering students at the Stevens Institute of Technology to fit
on the Franklin fuselage and use the same struts. The increase
36
in control and performance is very marked, the gliding ratio being

about 17 to i and the sinking speed under 3 feet per second,
bringing this glider almost into the intermediate sailplane class.
The cost of a utility glider is about $600 new, including a trailer
for transporting it disassembled. Some firms have put out partlyassembled kits for a price of about half this amount. Used sec
ondaries or utilities in good condition and with trailers can some
times be purchased for $350 upwards.
THE INTERMEDIATE SAILPLANE
The intermediate or training sailplane is a somewhat recent
development to fill the gap between the secondary glider and the
high performance sailplane. Due to its exceedingly low aerodynamic
drag and its very flat gliding angle the latter type is often difficult
to fly for the student trained in the former. The intermediate sail
plane fills this gap perfectly so that the transition is more gradual
and consequently safer.
The chief differences between this type and the secondary are
its greater span with higher aspect ratio, higher cruising speed and
generally cleaner design. The span of the average intermediate
sailplane is 45-48 feet and the aspect ratio 14. Built to stand the
stresses of soaring in rough air and aerobatics, its weight is about
290 pounds and wing area 170 square feet, giving a wing loading
in flight of about 3 pounds per square foot.
The first reaction of a student after flying an intermediate sail
plane is its "slippery" feel due to its clean design and resultant
low drag. Its gliding ratio is about 20 to i and sinking speed about
2.8 feet per second. Stalling speed is 26-30 m.p.h.
Cost of an intermediae sailplane when new is about $750. This
is the price of the Bowlus "Baby Albatross" which can also be pur
chased in kit form for $425 available in separate units, the first
costing $75 and each subsequent unit $35. An excellent American
intermediate is the ABC Sailplane, designed and built by Arthur B.
Schultz, President of the Detroit Glider Council. It won the Eaton
Design Competition for the most practical American glider pro
duced in 1937. A set of detail working plans for this sailplane is
available at a cost of $35. There are still so few intermediate sail
planes that used ones are seldom on the market.
HIGH PERFORMANCE SAILPLANE
The most advanced type of sailplane, usually designated as high
performance, is the most beautiful and perhaps also the most effi
cient of all heavier-than-air aircraft. This type a few years ago was
THE BOWLUS-DU PONT "ALBATROSS" HIGH PERFORMANCE

SAILPLANE OF 1934
I-rank Turgeon, Jr.
GOPPINGEN I "\VOLF"
INTERMEDIATE
SAILPLANE
Ham Groenhoff
THE Ross "!BIS" HIGH PERFORMANCE SAILPLANE TAKING OFF
Hans
38
characterized by large wing spans of more than Go feet. An example

was the Bowlus-du Pont "Albatross" which measured 62 feet from
tip to tip of its highly tapered wings. The sailplane was typical of
the best designs at that time, none of which were quite so clean or
nearly so strong as modern high performance sailplanes. Although
an extreme limit of 98 feet span was reached in the "Austria,"
spans have now come down in the interest of greater maneuver
ability necessary for efficient spiraling in thermal upcurrents and
ease of handling on the ground. They vary from the 48 feet of the
American Ross "Ibis" to 50 and 56 respectively for the German
"Rhonsperber" and "Minimoa" designs. Aspect ratios average about
16 to i, although some experimental German designs have been as
great as 30 to i. Weight of these sailplanes will vary from the 310
pounds of the small Ross to 520 pounds for the 1937 model of the
"Minimoa." Wing areas vary from 125 to 205 square feet for the
two types. Wing loadings are from 31/2 to 4 pounds per square foot.
The most outstanding of the flying characteristics of the high
performance type is its extraordinarily flat gliding ratio. The Ross
and "Minimoa" are both about 26 to i. Experimental German
types have gone as high as 33 to i. This flat glide coupled with a
low sinking speed of 2 feet per second naturally gives this sailplane
excellent climbing ability. At a soaring contest where a number of
gliders of different types are slope soaring the high performance
sailplanes soon climb 500 or 1000 feet higher than the secondaries.
Other important flying characteristics of this type are its cruising
speed of about 40 m.p.h. which is 2 to 5 miles above the stalling
speed, and its efficient speed range. The best of these sailplanes have
a sinking speed of just under 2 feet per second at the cruising speed
of 40 m.p.h. The clean design helps it to fly quickly through the
downdrafts without much loss of altitude.
On the author's distance record flight with a "Minimoa" the 212
miles were covered at an average speed, measured for the airline
distance, of 37 m.p.h. There was an average tail wind of about
25 m.p.h., but when there is also taken into consideration the facts
that the actual course covered was at least 260 miles and that during
the greater percentage of the time the sailplane was being spiraled,
its speed performance becomes more apparent. The world distance
record of 405 miles was made in 8|4 hours, an average of close to
50 m.p.h.
In keeping with the general streamlining of this type, the pilot's
cockpit is entirely enclosed. As well as helping the streamlining it
also often is a necessity for pilot protection from airflow at high
speeds or from cold at high altitudes. On a flight made in the "Ibis"
TYPES OF GLIDERS
39
in New Hampshire the temperature was 12 F. at 9500 feet altitude,

but due to the fact that the ship was soaring in the sunlight above
the clouds the inside cockpit temperature was a comfortable 75 F.
On the world.record altitude flight to over 26,000 feet above sea
level a temperature of 40 below zero F. was experienced.
Hans Groenhoff
THE SCHWEIZER TWO-PLACE, ALL METAL, HIGH PERFORMANCE

SAILPLANE
Due to its size and complexity of design and construction the
high performance sailplane is the most expensive of gliders. Average
prices usually vary from $1200 to $3500.
Several two-seater sailplanes have been built, falling in the last
three categories. Their performances are the best criterion of the
proper classification. An outstanding sailplane of this type is the
all-metal Schweizer sailplane which combines high performance with
the ease of handling of a utility. Specifications of this ship, with
which several national records have been made, are as follows:
span, 52 ft.; aspect ratio, 12.6; weight empty, 450 Ibs.; wing area,
214 sq. ft.; gliding ratio, 23 to i.
CHAPTER IV
DESIGN, CONSTRUCTION AN.D

MAINTENANCE
By Paul and Ernest Schweizer
I HIS DISCUSSION OF glider and sailplane design and con

struction is not intended to be a text on the subject. Glider design
is essentially a branch of aeronautical engineering and requires
technical knowledge, as well as knowledge of gliding, on the part
of the designer. Many gliders have been built without any formal
technical knowledge by men who learned from practical experience
and by cut-and-try methods. This type of work has enabled many to
fly gliders but on the other hand many others were discouraged
because of the poor results obtained and the accidents which hap
pened rather frequently. Persons with limited technical knowledge
who wish to design and build gliders should at least have the advice
and supervision of someone with this necessary knowledge. Lack
of supervision in building and operating gliders brought on the
rigid regulations now in force. It is much more satisfactory for all
concerned that a good set of plans be followed exactly. This dis
cussion is intended to cover the various types of construction, to
give general comparisons and to point out the problems the designer
must consider and work out in his designs.
LOADS AND STRESSES ON STRUCTURE
The strength and design requirements for gliders and sailplanes
are set by the Civil Aeronautics Administration of the federal gov
ernment and they are based upon past experience and research in
this country and abroad. It is obvious that the different types of
gliders and sailplanes require different strengths. The strength fac
tors used for training gliders and sailplanes are much different from
those used for cloud flying or acrobatic sailplanes. To take care of
these different factors, the requirements are based upon flight, gust,
and towing speeds instead of having special requirements for each
particular type. This method allows the designer to pick his speeds
to suit the purpose of his ship more closely. The glider is then
placarded with these speeds as limiting speeds which should not be
exceeded.
40
CONSTRUCTION AND MAINTENANCE
41
The strength and design requirements represent the minimum

strength, etc., necessary for safe operation under given conditions.
It is obviously impossible and certainly impractical to design for
all possible conditions in which gliders and sailplanes might fly.
Nature is capable of an infinite variety of releases of atmospheric
energy in which an aircraft may find itself. It is up to the pilot's
experience and discretion, and quite often up to chance, to avoid
conditions that exceed the strength limits of his plane. Safety factors
and extra design margins take care of many unusual conditions but
it is impossible to take care of all.
There are three types of loads that act on gliders and sailplanes
in flight: gust loads, maneuvering loads, and towing loads. Gust
loads are due to sudden changes of speed or direction of the air in
which the plane is flying. Maneuvering loads are caused by con
trolling the plane in flight. Towing loads occur in winch, automo
bile or airplane towed flight. These three types of loads may act
separately or in a variety of combinations.
It is quite apparent that sailplanes in search of upcurrents will
encounter gusts, as an upcurrent is essentially a gust of rising air.
These sudden accelerations, possible in all directions, build up loads
in the structure of the plane entering them proportional to the gust
speed, the flight speed, and the sharpness of the gust. The strength
of gusts to be expected for different conditions has been determined
from experience and checked by flying through them with recording
instruments. Through logical formulas, the strength required to
resist these gusts can be determined for any speed of gust and flight.
In most cases it is the gust loads that determine the designs of the
wing and the supporting structure.
In normal gliding and soaring, the maneuvering loads are consid
erably less than the gust loads. The loads in sharply-banked turns
or in an occasional sharp pull-up are about the maximum maneu
vering loads normally experienced. In cloud and storm flying the
maneuvering loads may be high because of the necessity of adequate
control in rough air at high speeds. But here again the gust condi
tion is the design condition as the gusts in storm clouds are ex
tremely strong and sharp edged. Gliders designed for stunting and
unlimited aerobatics will almost always be designed by the maneu
vering loads. It is evident that severe strains can be put on a ship
in this type of flying.
Towing loads are determined to a large degree by the method
of towing used. Airplane towing imposes the smallest loads of any
of the various methods. In the average towing condition the load on
the towline is approximately the drag of the sailplane. The danger
42
in airplane towing does not lie in the actual towing loads or forces
but in the speed at which the glider is being towed. If it hits a gust
at this speed the loads on the glider are naturally greater than at
normal gliding speeds. Also if any sharp maneuvers are made while
in tow the loads will quickly build up.
In auto and winch towing the loads are large for ordinary con
ditions and can easily become excessive in windy and gusty weather.
The wings of the glider support not only the weight of the glider,
but also the load on the towrope and the balancing load on the tail.
The intensity of load on the towrope has actually been measured
in flight and during a tow of a utility was found to be as high as
500 pounds. Apparently it often goes above this as towropes of
greater strength frequently break in towing. It is the practice to
limit the loads on the towrope by putting a weak link in the line.
This will allow the rope to break before excessive loads can be put
on the ship. In most cases the gust condition will cover towing con
ditions but the towing speeds should be limited to keep within this
range. The nose of the fuselage has to be designed to carry the
rope loads which may sometimes act almost vertically.
Shock-cord launching, although not imposing great loads on the
wing, does require adequate strength in the fuselage. As the shockcord loads may easily go over 1000 pounds, it can readily be seen
that the fuselage between the nose-hook and hold-back point has
to be quite strong. This load more than takes care of the tension
loads (forward) of the other types of towing.
As these various types of flight loads can occur separately or in
various combinations, their combined effects also must be investi
gated. Towing in gusty weather with excessive maneuvers necessary
to keep the ship in line represents a combination of all three that
can very easily be encountered. Although it is poor practice to fly
in very gusty weather these conditions do occur and should be con
sidered in design.
All loads considered so far have been flight loads. Although these
are of primary importance they are not the only design conditions.
Landing, handling, and crash loads have to be considered also if a
practical ship is to result.
Handling loads play an important part in the design of gliders
and sailplanes, because of the general lightness of construction. In
many places the strength required to meet the air loads is not suffi
cient to take care of the handling loads. The wing tips, although
taking a very small air load, have to be rugged enough to stand the
strain of handling and normal tip landings. The fairing members
in many places have to be reinforced to stand up under the tight-
43
ening effect of dope. Many other parts would be too fragile to

handle if they were designed only for the air loads.
The requirements for landing strength are more severe for train
ing ships than for sailplanes because the training ships are often
handled by students or inexperienced pilots. There are definite
requirements for the different types of landing conditions that
should be designated for. It is good practice to make the ship rugged
in this respect for there is always a chance for a hard landing in
any type of ship.
Although crash loads are difficult to determine and as there is
much debate as to how far one should go in strengthening against
these loads, it is apparent that some minimum requirements of
strength and ruggedness are necessary to guard against injury from
ordinary crack-ups. Stalled landings, head-on crack-ups, nose-over
landings, etc., are some of the conditions that should be designed
for. The extra weight used to strengthen against these crack-ups will
be well worth while and this strengthening will give the pilot a
better feeling of security.
Fuselage
The fuselage is essentially a structural member in the glider. It
has no important beneficial aerodynamic purpose. The fuselage sup
ports the pilot and the tail surfaces and transmits their loads to
the wing structure. It also transmits all the loads to the landing
gear in landings. These facts are the prime considerations in the
fuselage design and construction. In the design of any particular
fuselage they are considered along with other factors such as cost
of construction, ease of repairing and maintaining, low aerodynamic
drag, pilot's comfort and convenience and many detail considera
tions.
The primary fuselage is the simplest form of fuselage that will
serve all the structural purposes. It is cheap and simple to con
struct. The construction takes the form of a simple frame structure
which has all its members exposed to the air. Aerodynamic consid
erations are not very important. The material used may be either
steel tubing or plywood and wood. The methods of assembly are
conventional to each particular type of construction.
In all other types of gliders the aerodynamic drag is important.
The importance of drag increases progressively from the secondary
to the high performance sailplane. The drag is to be considered in
the choice of type of construction. The trussed framework or box
type is inferior to the stressed-skin or monocoque type because it
cannot be made to conform to the pilot's shape as effectively as the
44
monocoque type, resulting in a larger cross-section area. The monocoque type can also be made to approach the best streamlined form
and have minimum skin friction. Hence for sailplanes of the highest
performance class the monocoque type is used.
It is well to keep in mind the relative importance of the fuselage
drag. The difference in drag between a well-faired frame fuselage
and a streamlined monocoque fuselage is not very great and the
fuselage drag is only part of the total drag of the glider. For this
reason the frame type fuselage can be used successfully on inter
mediate and high performance as well as on secondaries and utili
ty WOOD COVERING
LONGEKONS
BOX TYPE OF CONSTRUCTION USED
CHIEFLY ON INTERMEDIATE
FITTING
TOP PLYWOOD
PLYWOOD
(SPRUCE Of PIHC.)
LONGERON
SIDE
FRAME.
f>*uce on fine.)
V LON&tRONS
(SPRUCE Oft PINS,)
CONSTRUCTION DETAIL
TYPICAL MAIN FRAME.
LDNOtKONS
TYPICAL MAIN FRAME
WOOD AND PLYWOOD Box TYPE FUSELAGE CONSTRUCTION

ties. For utility and secondary fuselages, simple fairing is sufficient.
It is more important to have a clean fuselage without exposed struts,
projections or openings, if possible, than to take a great deal of care
with the fairing. Also because the monocoque type is usually less
sturdy and more difficult to repair it is used principally on high
performance sailplanes where it will receive more careful handling
by experienced pilots.
The frame type can be built either in wood or metal. In European
practice, wooden framework structure is used with plywood cover
ing. There are no true fairing members, the sides are flat with a
"V" bottom and top to form a diamond-shaped cross-section. This
type is used with various minor changes such as rounded tops and
45
different types of cabanes for wing attachment. The plywood cover

ing carries some stress as tension load in the framework bays, as
no diagonals are used. The frames are built of pine, spruce or ash
and assembled with plywood gussets. The longerons are also of
spruce, pine or ash. Birch, mahogany and spruce plywood are the
most commonly used. Steel fittings are bolted to the frames with
proper reinforcements to carry the loads through. If fabric covering
is used, some fairing must be provided and the bracing of the struc
ture must be complete. Aluminum alloy also can be used, although
not much has been done with it in frame type gliders. Aluminum
alloy shapes or tubes can be used as frames with aluminum sheet
RUDDCR
REMOVABLE COWL
FAIRING TUBCS
CABLE. AND PUSH-PULL

CONTROL SYSTEM
SINGLE. VJHttL
LOW PRESSURE. TIRE.
TAIL SKID
RUBBER SPRUNG
FUSELAGE STRUCTURE
WELDED STEEL TUBING
WIN& STRUCTURE EITHER

MOHOfPAK OK
STRUTS
I-SPAR SrRI/T-8ff/ICED
WING TIP SKID

FOR TRAINING
STEEL OK ALUMINUM ALLOY

STREAMLINE TVBINb
UTILITY GLIDER OF WELDED STEEL TUBE CONSTRUCTION

or fabric covering. The fastening must be limited to rivets, bolts
and screws, as welding is not applicable.
In American practice some use is made of the wooden types of
construction, but in general the welded steel tube fuselage is pre
ferred for utilities and medium performance sailplanes. The struc
ture can be made more simple than the wooden frame type and
more efficient because of the better joints that are obtainable by
welding. The steel tubing is very durable and easily repaired.
Fabric covering is used and fairing members may be made of metal
tubes or formed pieces, or wood strips. The minimum sizes of tubes
specified for airplane practice do not apply but the effect of hand
ling loads, etc., must be considered. The welder must be properly
qualified to build this type of fuselage. The proper design of joints
and fittings requires some experience in this type of design. Fittings
46
and control bearings, etc., are welded directly to the structure. Three
or four longeron type structures or combinations of both, which
may suit some designs better, are used. Standard airplane prac
tices of protecting the tubing internally and externally are recom
mended.
A monocoque fuselage can be made to conform closely to the
pilot's outline since there are no large structural members to inter
fere and the shape of the cross-section does not depend on any
system of framework. The plywood monocoque is the type most
BULKHEADS
TyPICAL
SAILPLANE
TAIL SKID
FUSELAGE
(SfMI-MONOCOQUC SINCE A FEW LIGHT

STRINGERS OR LONGERONS ARE USED)
PLYWOOD
COVERING
TYPICAL
CONSTRUCTION
PLYWOOD SHELL TYPE FUSELAGE CONSTRUCTION

SEMI-MONOCOQUE
commonly used in sailplane construction. Actually it is a semimonocoque type since it always has at least three longerons or
stringers. Frames or bulkheads are used to carry direct loads from
the wings, landing gear and concentrated weights to the shell. Some
are also used in the rear to carry the tail surfaces. Intermediate
frames are made lighter as their main function is to support the
shell and give the fuselage its form. Part of the efficiency of the
structure is due to the fact that the shell acts as both structure and
fairing. The plywood covering is applied to the frames in a series
of lateral bands or rings.
On the front part of the fuselage several light stringers and longi-
47
tudinal strips of plywood may be used because the smaller radius

of the fuselage lines will not permit conformity of the lateral rings.
Scarf joints are used in the plywood to maintain a smooth surface.
The nose piece is usually carved of wood or formed from sheet
aluminum. By careful sanding and by using a rubbed paint or
varnish finish, an extremely smooth surface can be obtained. In
construction, the alignment of the structure is maintained by
some form of jigs. Frames and bulkheads are built up of wood and
plywood. Severely formed parts are laminated. Spruce, pine, birch
and ash may be used for both the frames and the stringers, depend
ing upon the designer's judgment and the materials readily avail
able. Ash and birch are used where considerable strength and
hardness are desired.
The metal monocoque fuselage has not been very extensively used
in sailplanes but is rapidly becoming popular because of its strength
and durability under difficult conditions and because it lends itself
to forming operations more readily than plywood. Most of the types
of monocoque metal construction used in powered aircraft can be
applied to gliders, depending on what is desired of the fuselage.
If a very light structure is desired, the use of light skin with numer
ous stringers and frames will be most suitable. This type of struc
ture is more complicated since it requires a large number of parts
and riveted joints. The true monocoque with heavier skin and
no stringers shows promise of being very economical to build and
maintain, as \vell as being very rugged. This type is likely to be
somewhat heavier than the other. When reasonable production is
possible this type of construction may be applied economically to
ships of the lower performance class.
The boom type of fuselage which uses a metal tube for the rear
part of the structure has found some application. This type has the
advantage of considerably reduced friction drag. The metal tube
itself is quite economical, but problems of simple and satisfactory
attachment, and flutter may reduce its overall efficiency and
economy. If a simple tube is used its section cannot easily be varied
to get the maximum efficiency. The wall thickness must be relatively
heavy to obtain stiffness. Tubes fabricated from sheet metal or wood
are not likely to be efficient.
Wings
The complete design of a wing is a rather complex problem. It
is necessary to consider aerodynamic characteristics, weight of struc
ture, type of structure, material, ease of construction, maintenance
and purpose. In general the designer avoids this complexity as much
48
as possible by drawing on contemporary and previous practice. New

gliders are rarely designed without regard for previous practice
unless they are intended to serve some special or experimental pur
pose. The usual design is a modification of previous practice for
the purpose of increasing the performance and utility, reducing the
cost, or serving some special purpose. The aerodynamic design of
wings is closely tied up with the structural design which definitely
limits the performance obtainable.
The design of wings generally can be grouped according to the
purpose for which the glider or sailplane is to be used. The group
ing below indicates the usual range of construction and it is appar
ent that the designer has considerable leeway.
Primaries: General2-spar, wire-braced; Occasionally2-spar;
strut-braced; Rare or Nevermonospar, cantilever.
Secondaries and Utilities: Generali- or 2-spar, strut-braced:
Occasionally2-spar, wire-braced, monospar; Rare or Never
cantilever.
Intermediate Sailplanes: Generali- or 2-spar, strut-braced; Occa
sionallymonospar, cantilever; Rare or Neverwire-bracing.
Sailplanes (high performance): Generalmonospar, cantilever and
strut-braced monospar; Occasionally2-spar braced; Rare or
Neverwire bracing.
There are two principal classes of wing structures used in the
construction of gliders and sailplanes: the monospar and the twospar types. Multispar wings have rarely been used on gliders but
might possibly be used on high performance sailplanes in metal
construction. The monospar type of construction until quite re
cently has been almost unique to gliders. The two-spar wing is sub
stantially the same as that used in other aircraft. Both one- and
two-spar types have several variations according to type of bracing
and material used.
There are general principles that apply to all types of wing con
struction. The wing structure must be able to withstand all types
of forces that can be applied to it in its intended range of opera
tions. Wings are not designed to be indestructible but are designed
to withstand varying design loads which are determined from the
purpose of the ship. The reason for using different kinds of structure
is that a number of varying types of loadings exist, the proportion
and magnitude of which, along with aerodynamic and cost consid
erations, determine which type is the most suitable.
The air forces on a wing are divided into three parts: the lifting
or normal forces, the drag forces and the twisting moments about
some axis along the span. In the monospar wing the lifting forces
49
are taken by a single spar or beam. In some cases this beam may
also take all the chord loads and the twisting moments. This method
has been used in some airplane designs. The more common method
is to use the spar and covering forward of the spar to form a D-tube
which is resistant both to bending in the plane of the chord and
to torsion. The ribs and skin of the wing serve the purpose of giving
it the proper profile, to obtain the desired aerodynamic character
istics, and to transmit the air forces to the main structure. If they
can be modified to serve also as structural members, greater struc-
\x
TVPICAL
SPAR Bilffî
FITTINGS
_L^ !
1 SPAR
DRAG *TRUS\
Mt?S
DRAG TRUSS WIRES

AIRCRAFT HARD MIRE
tV/A/6
_|_ _
COMPRESSION
STRUTS 1
1
UTILITY
'
PLAN
!,.,,!
VIEW
! ^!^SsN\iV( ' r\\ ^5v

">
STRUT\ /
FITTING^
X7
/X
t
>ING TIP
BOH
/\/\/\/\l\/*>
^- AILERON
^TRAILING
OGE
COMPRESSION STRUT
LMOW& EDGE
COVERING
REAR SPAR
RIBS OF SPRUCE
** PLYW00D
FRONT SPAR.
SPflRS StllD SfRI/CE <* -/- SfCTION
AIRCRAFT BOLTS
SPAR PIN
SPAR BUTT FITTING

AIRCMFT STEEL
SPAR
WING
DETAILS
., '*^-STRUT
DETAILS OF TYPICAL 2-SPAR WING

tural efficiency is obtained. The D-tube method is used with either
metal or wood construction. Another method is to use an auxiliary
spar. A torsion and drag-resisting truss is built between the main
spar and the auxiliary spar. This type of construction is used on
powered aircraft and may find some application in gliders, especially
if the rib structure can be combined with it.
In the two-spar wing the torsion loads are resolved into two loads
which are applied at the front and the rear spar. The drag forces
are carried between the two spars by a truss, called the drag truss.
The two-spar wing is easy to build and design and is quite efficient
when used on smaller spans or on relatively thin wings when large
torsional loads are present. Because the spars are placed a consider-
50
able distance apart, the full depth of the airfoil section cannot be
used. These two spars are never working at full efficiency at the
same time because of the wide range of center of pressure travel
along the chord. In two-spar wings of high aspect ratio it is neces
sary to use some additional means of torsion bracing, such as a
double-drag truss, to give sufficient torsional rigidity. Because of
this the monospar wing is used more for high aspect ratios.
The normal loads of the wing may be carried by a cantilever
beam or by some form of a trussed beam using either struts or wires.
The wire-braced type is used chiefly on primaries and secondaries
because it is light, simple, and economical to build. However, this
type is inconvenient to set up and rig and has high drag. The strutbraced or semi-cantilever type is used on secondaries, utilities and
intermediate sailplanes and also on some high performance sail
planes. Streamlined struts are used: one strut on the monospar wing
and two struts on the two-spar wing. There are exceptions to this,
however, as two struts in the form of a "V" are sometimes used to
carry torsion from a monospar wing, and one strut can be used
on a two-spar wing in certain cases.
For utilities and secondaries the convenience of setting up a strutbraced ship more than compensates for any additional weight or
cost. There is considerable variation in the design of strut-bracing.
With the relatively thin airfoils (about 12%) used on utilities, long
struts with short overhangs are used to obtain low bending moments
in the spars. This requires the use of jury struts to keep the weight
of the struts down. For monospar wings a slight increase in thick
ness permits the use of a longer overhang and a shorter strut with
out a jury. This gives about the ultimate simplicity in setting up,
and efficiency as well. The same general design is used on many
intermediate sailplanes and on some high performance sailplanes.
On normal sailplanes the use of a considerably thinner airfoil for
the braced monospar than for the cantilever wing gives both types
about equal efficiency. The weight difference between the two is
slight. The struts are subjected to high compression stresses and
are quite heavy unless bulky sections are used. Additional fittings are
also required for the struts. For general convenience in construction
and operation struts are used on practically all the intermediate
sailplanes while both struts and cantilever are almost equally used
on high performance sailplanes. Sailplanes designed for high cruis
ing speeds may use thin, high aspect ratio, cantilever wings. The
high wing loadings of this type permit the use of heavy construction
necessary for thin wings.
Biplane gliders have not been used extensively. Early training
51
types were sometimes biplanes because very low wing loadings could
be obtained with relatively small spans. Modern biplane gliders are
rare. The biplane has a theoretical advantage because for a given
span a biplane may have less induced drag than a monoplane. The
bracing system necessary for a biplane usually will have more drag
than the monoplane bracing. However, the biplane may also be
built with cantilever wing construction. In general there are no
serious limitations to the span of gliders and the added complica
tions of attaching and constructing two wings do not favor the
development of the biplane. The construction of the wings may be
of wood or metal with the usual variations of design.
LEADING DG J/T//V n CARRY
DRAG- AND TORSION LOAC3
AILERON
FITTINGDRAG AND TORSION MEMBER
TYPICAL
M ON OS PAR
PLAN
WING
V/W
WOOD
O1QE
MONOSPAR
WING
CONSTRUCTION
ALLOY
ALUMINUM
MONOSPAR
CONSTRUCTION
MONOSPAR WING CONSTRUCTION

There are numerous plan forms used on glider wings, the choice
again depending upon the purpose of the design. The elliptical
wing has the best aerodynamic efficiency because it has the minimum
induced drag. For sailplanes the elliptical cantilever wing has been
used extensively. A more thorough study shows that a tapered wing
gives better overall efficiency, the result of combined aerodynamic
and structural efficiency. A third type using straight center sections
and tapered tips gives nearly the same overall efficiency as the
tapered wing and has the advantage of greater simplicity of con
struction. It is widely used on standard type sailplanes and is also
the most frequently used type on intermediate sailplanes.
In some designs the plan form is modified by increasing the chord
of the ailerons to obtain better control at the cost of some efficiency.
The rectangular wing is used on primaries because it is the easiest
to construct, The tips of the rectangle are usually modified, as this
52
can be done quite readily. Secondaries and utilities also use the
straight wing but higher aspect ratios are used and more care is
taken with the tips to obtain better performance.
The choice of plan form is influenced to a greater degree by
structural and practical considerations than by aerodynamic effi
ciency. Well established data show that at an aspect ratio of about
7, as commonly used on utilities, the difference in induced drag
between the rectangular wing and the elliptical wing is less than
60% and the resulting difference in drag for the total glider would
be considerably less. Moreover, this difference can be reduced fur
ther by modifying the shape of the tips and by using some "wash
out," which would leave the elliptical wing an advantage of about
2 % or 3% in aerodynamic efficiency. The effect would be about
twice as great for aspect ratios of 16 to 20. Tapered wings and
wings with straight center sections and tapered tips closely approach
the efficiency of the elliptical wing. Thus, as pointed out before,
the structural and practical considerations will easily outweigh the
rather small aerodynamical advantages of the elliptical wing.
The value of gull wing design has caused much discussion. Its
main feature is that it gives increased wing-tip ground clearance
without using dihedral over the entire span. From a structural
viewpoint it is rather an undesirable complication. Its advantages
for stability cannot possibly be great and the current trend of wing
design has veered away from the gull wing form to that of simple
dihedral.
In detail design the two-spar wing is similar to the airplane wing.
In general it is lighter and simpler in construction. The spars may
be of either wood or metal; solid wood for lower performance types
and I-beams or box beams for the higher performance types. It is
permissible to use deeper and thinner spars than in airplanes. The
drag truss may be built up of either wood or hard-wire bracing.
A system of double wood bracing is used on some designs, giving a
very rigid wing. The hard wires may also be used double to improve
the torsional rigidity of the wing. The ribs are usually of the truss
type using spruce with plywood gussets. Plywood web types are
rather heavy and expensive and are not extensively used. Stamped
aluminum ribs are light and economical to produce commercially
because less labor is required. Welded steel ribs of special design
also have been used on some gliders. Leading edges usually are
covered with plywood or sheet aluminum for about 10% of the
chord; or false ribs and a leading edge former may be used.
Aircraft carbon steel and chrome-molvbdenum steel are used for
53
fittings. Duraluminum is also used where extreme lightness is de

sired but is not as satisfactory as steel if subjected to hard use.
Airplane practice should be followed in the use of hard wires,
cables, bolts, shackles, turnbuckles and other standard parts. Wood
and plywood should be of aircraft specification. Species of wood
other than spruce may be used if of proper quality, particularly
on training ships where very low cost is desired. High strength
aluminum alloys are reliable and of good quality if obtained from
LEADING EDtE. COVERING

OF PLYWOOD
LC. RIB
SPRUCE.
PLYWOOD
SPAR \fJEB3
+5' PLYWOOD
SPAR FLANGES
SPRUCE.
ALL JOINTS WOE

~,r, AIRCRAFT CLVE
PLYWOOD RIB
ATTACHMENT &VSSET
REAP ftia or 'SPRUCE
> PLYWOOD
ALUMINUM ALLOY MONOSPflR

CONSTRUCTION DETAIL
WOOD MONOSPAR CONSTRUCTION DETAIL
DETAILS OF METAL AND WOOD MONOSPARS

proper sources. The use of commercial steels for fittings is not of
sufficient advantage to make it worth risking. Good aircraft steels
are reasonably priced and the quality required for the average
glider is quite small.
Plywood is an important factor in the monospar wood wing. The
spar is invariably either a box or I-beam. The ribs are made in two
sections, the front and rear, as the spar is made the full depth of
the airfoil. The rear ribs are cantilever, transmitting their loads
directly to the spar. An auxiliary spar may be used to support the
ailerons and to take the rear drag fitting, but not to take any beam
loads. The torsion and drag loads may be taken by a fitting at the
leading edge or by a rear drag fitting on the auxiliary spar or
54
diagonal brace. The D-tube is built up by covering the front section

with plywood glued to the ribs and spar with casein glue. A
light wood stringer is used in the leading edge to support the ply
wood. In some designs more stringers are used with thinner ply
wood. The D-tube does not permit direct accurate mathematical
analysis of its strength. The designer must have some data upon
which to base the design. Structural tests are necessary to determine
the strength of the complete structure.
The D-tube, monospar construction lends itself well to metal
construction. Stamped aluminum ribs are used with aluminum
alloy sheet instead of plywood. The spar can be constructed of
sheet for the web and extruded sections for the flanges as in air
plane practice. Joints are made with rivets, screws or bolts, depend
ing on the location. The self-tapping sheet metal screws are ex
tremely useful as they give strong joints that can be opened easily
for repairs. They are also very valuable in places where it is difficult
to buck rivets. For very lightly loaded wings, the wood wings may
have a slight weight advantage because the low density of plywood
makes it relatively stiff for its weight. In structures that are more
heavily loaded, the greater density of the metal permits it to be
used more efficiently than the wood. Because of its many advantages,
metal construction is rapidly becoming popular, as it is in airplane
construction.
Ailerons
The ailerons are more uniform in design than other major parts
of gliders and sailplanes. A much larger percentage of aileron area
is used than on airplanes. Primaries use simple ailerons without
balancing or differential action. On all other types some form of
differential is used between the two ailerons. Maneuverability is a
very important factor in soaring flight and it is desirable to obtain
it with a minimum of loss in efficiency. The simple aileron has high
adverse yawing (turning) moments, especially at high angles of
attack. This requires the use ot more rudder control and gives
sluggish and inefficient turns. The differential action aileron tends
to reduce the adverse yawing moment.
Rigging up ailerons also helps to improve the turns. Extreme
differential (using up movement only) is very effective. On sail
planes the percentage of aileron chord is usually increased toward
the tip and in some designs the ailerons are extended behind the
normal trailing edge line. Various types of slotted and Frise ailerons
exist but are not used extensively on sailplanes. Static and aero
dynamic balance are used in a few designs but do not seem to be
55
necessary except in some of the larger designs. It is, however,

advantageous to keep the control forces as low as possible to avoid
pilot fatigue and to make the controls responsive to a light touch.
The aileron gap requires some sort of fairing as an open gap will
cause drag and inefficient operation of the ailerons. This of course
does not apply to the slotted or Frise type ailerons which require
a definitely proportioned gap.
In wings of large span two control horns are used on the ailerons
to reduce twisting loads. If the wing structure is likely to be rather
flexible the ailerons are made in two separate parts to avoid binding
of the hinges. The two parts may move together or may have a
differential action between them.
An auxiliary false spar is frequently used to carry the aileron
hinges and the necessary gap fairing. In the two-spar wing construc
tion, the proportion of the aileron and the position of the spar
often make the false spar unnecessary. Some monospar designs use
cantilever hinge brackets to carry the aileron loads to the main spar
and D-tube. The false spar may be supported by reinforced ribs or
by special brackets.
Two types of aileron construction are commonly used. In one
type the ribs carry the only direct aerodynamic load to the torsion
member. The control horn is mounted directly on the torsion
member which also acts as the beam to carry all the shear loads to
the hinges. The other type uses a spar which is not resistant to
torsion, to carry the shear loads. The ribs are arranged to form a
truss to carry through the torsion loads to the control horn. The
first type is adaptable to long narrow ailerons and to balanced
ailerons of the slotted and Frise type where it is desirable to: con
centrate the weight of the aileron near the front. The second type
gives a lighter structure and is used more extensively. The mate
rials, as in the rest of the wing structure, may be of wood, plywood
or metal. Various other types exist, such as metal or plywood box
types and others with different type of torsion bracing and derails.
Hinges may be of the conventional airplane type. Aluminum alloy
piano hinges can be used to advantage in eliminating the gap, the
hinge being used on the upper surface of the wing.
Rotating wing tip ailerons have been used on some gliders but
are rarely used on any new designs. These ailerons are adapted.to
sailplanes with long tapered tips. Installation is made directly on
the control torque tube which also acts as a cantilever beam. The
tip aileron is easily susceptible to damage on the ground and in
general does not compare with the conventional type aileron in
efficiency and effectiveness.
56
Wing Fuselage Connections

In seaplane design the wing and fuselage connection is rather
important as it is a great source of drag and disturbance if not
designed properly. The earliest types of sailplanes used the high
wing exclusively to permit more depth for the strut bracing and
to keep the wings away from the ground. When the wing is just set
on the fuselage the interference drag is high. The connection is
more efficient when the high wing is set on a neck or cabane ex
tending up from the fuselage. Cabanes consisting of a number of
small supporting struts have considerably more drag than the full
fuselage cabane.
Numerous tests in wind tunnels have shown that the midwing
and shoulder wing installation have the least interference drag.
Sufficient depth for struts can usually be obtained with the shoulder
wing and a much more rigid fastening of the wing to the fuselage
is obtained with the mid- or shoulder wing than with the high
wing. The fittings can be fastened directly to the large parts of the
frames or bulkheads, giving greater strength.
Mid-wing and shoulder-wing types are more likely to cause
burbling and buffeting of the tail surfaces, and fillets are usually
required at the junction of the wing and fuselage. The exact type
of filleting required can be determined accurately beforehand only
by wind tunnel tests. There are, however, considerable data avail
able which will give the designer a good idea of the best design.
Tail Surfaces
The tail surfaces of gliders usually are conventional with many
minor variations in design used by individual designers. On normal
ships a stabilizer and fin are used, but in the past hone were used
on high performance sailplanes and in some cases even on training
ships. In the case of the sailplane the undamped surfaces (those
without fixed stabilizer areas) were used to obtain very responsive
controls but in training gliders the only advantage was ease of con
struction. It is now the general practice to use a stabilizer and
some fin area on practically all sailplanes and gliders. Higher speeds
and blind flying require more stability and make this necessary.
Adequate maneuverability is obtained on sailplanes with a stabilizer
and it is required for safety on training ships. The fin area is of
lesser importance as the fuselage has some fin effect, but it is re
quired in some designs to obtain the desired stability.
The most important requirements of the tail surfaces arc that
they give satisfactory maneuverability and stability with the mini-
57
mum amount of drag. Ease of assembly is quite important and

construction must be considered. Cantilever surfaces are used on
practically all sailplanes while strut-braced surfaces are used on
most intermediate types. Wires and struts are used on primaries,
secondaries and utilities. Aerodynamic balance is frequently used
on rudders and on some elevators. Static balance has been applied
to a few of the larger sailplanes for elevators and rudders.
The undamped type of elevator most used is the balanced canti
lever type. It is made in two halves mounted on either end of the
torque tube running through the fuselage. The torque tube also
carries all the cantilever loads of the elevator. If this type is used
care should be taken to see that the torque tube and its bearings
are free from play and rugged enough to withstand reasonable
handling loads. The position of the torque tube must not be too
far back as aerodynamic overbalance can cause dangerous condi
tions. It might be repeated here that it is now considered good
practice to use a damped horizontal control surface on all gliders.
The arrangement of tail surfaces varies with designer and type of
glider. They are perhaps best discussed in groups. In sailplanes
external horns, wires, struts and fittings are to be avoided. It is
desirable to keep the torsion loads about the fuselage due to the
tail surfaces as low as possible. The horizontal surfaces are usually
made in two parts, permitting the rudder centroid to be near the
fuselage axis. The rudder area is also disposed so that the centroid
is as low as possible. The elevator axis is usually placed a short
distance ahead of the rudder axis permitting an internal horn on
the connecting torque member of the elevators. The stabilizer is
usually cantilever in construction, the rear spar carrying all the
beam loads and the resulting torsion loads being carried through
by the stabilizer ribs. The stabilizer leading edge member is usually
too shallow to act as a beam. In some designs a spar is used some
distance to the rear of the leading edge. Similar arrangements are
used in which the rear stabilizer spar is strut-braced. This is often
used in the intermediate sailplanes.
A new American version uses a joint in the elevator torque mem
ber in line with the pins of the stabilizer spar and leading edge.
This permits the horizontal tail surfaces to be folded up against
the fin by releasing the stabilizer strut pin. The construction of the
surfaces may be of wood, aluminum alloy or steel tubing. For canti
lever surfaces wood or aluminum alloy are best adapted. Steel tubing
is used extensively because of the ease of construction by welding,
but it is rather heavy. Remarkably light surfaces can be constructed
of aluminum alloy formed sections.
58
On primaries and secondaries the surfaces are usually wire-braced,

permitting simple sections, such as small tubing and solid wood
sections, to be used in the construction. Two separate elevators are
usually used and may be controlled by separate cables or by a torque
tube with the horn slightly offset from the fuselage. Various systems
have been devised to speed up assembly of the control surfaces,
particularly on utilities where quick assembly is important. In
some designs the fin and the rudder are removable and in others
they remain on the fuselage in trailing, the fin being arranged so
that the stabilizer is readily removable.
Another type that has found application on all types of gliders,
as well as on airplanes, has the fin and rudder set ahead of the
horizontal tail surfaces. This permits the use of a one-piece elevator
and simplifies construction and assembly of the tail surfaces. On
some sailplanes the resulting high position of the rudder is objec
tionable due to the large twisting moment on the fuselage. There
have been some failures of this type of rudder but they have been
due to 'poor detail design rather than to the design principles.
Cantilever or strut-braced types are used for the fixed surfaces as
the position of the fin and rudder is not convenient for wire bracing.
If the dimensions of the tail surfaces are kept low enough the surfa.!ces may be left installed on the fuselage for trailing. If the span
of the surfaces exceeds 7 feet, more care is required in trailing and
Oh longer trips it is safer to remove them.
In addition to the air loads, the tail surfaces are designed to
withstand handling loads, and some form of tail skid assembly
usually is necessary to protect the rudder from damage on landings
and take-offs. Wire bracing is susceptible to damage in heavy grass
or on landing in farm crops. Design loads for the stress analysis
will usually give smaller wire sizes than it is practical to use. In
strut-braced and cantilever surfaces care is necessary in designing
the assembly fittings so that there will be no play. Play will permit
vibration or flutter of the tail surfaces at high speeds or in some
condition in which a burble is set up by the wing or the fuselage.
Severe buffeting of the tail surfaces due to fuselage and wing
burbling must be corrected at its origin and not at the tail surfaces.
Large spoilers and flaps may also cause some vibration of the tail
surfaces. This-usually occurs only briefly and is not serious unless
it is very severe.
Control Systems
The control system is an essential part of any aircraft and in the
design of a glider it is important that the design and installation
59
of the controls be carefully considered. Reliability and ease of

control are the most important considerations. Simplicity in design
will do much to attain reliability and ease of operation.
Control cables are the simplest and most efficient means of trans
mitting control forces for most installations. For short lengths the
use of push-pull tubes or torque tubes is often more satisfactory.
For relatively long controls such as aileron controls, the push-pull
tubes will be heavier than cables, but considerations of quick con
nection in setting up, ease of installation and long life often favor
the use of tubes.
In order to attain reliability in the control system, only proper
quality of materials should be used. Aircraft standard cables, bolts,
pins, turnbuckles and other small parts should be used. Aircraft
sheet steel and tubing are also recommended for control systems.
It is advisable to follow good airplane practice in detail design.
This includes the selection of proper sizes of cables and tubes for
all installations. Hard wire has been used frequently for control
wire but this is very bad practice and should not be permitted.
Torque tubes should have sleeves at bearing points and at points
where horns are welded on, unless large margins of safety are pres
ent. All joints and pins require adequate bearing area for strength
and wear. Cable splicing should conform to aircraft practice and all
controls should have stops.
The detail design and installation of control systems vary so
widely that it is not within the scope of this chapter to cover them
all. In general the controls are similar to aircraft controls. Rudder
bars or pedals of various types and construction are used with cables
running directly to the rudder horns with as few pulleys and fairleads as possible. The stick control is used for most gliders except
that in some sailplanes the wheel or "dep" control is used. The
wheel permits the use of a narrower fuselage cross-section and in
large sailplanes with heavy control forces it is useful because of the
greater mechanical advantage that it permits.
The control stick is usually mounted on a fore-and-aft torque
tube which operates the ailerons. The stick is pivoted on the torque
tube for elevator movement. Cables or push-pull tubes are used to
transmit the control forces to the elevators. Another type of installa
tion has the torque tube across the fuselage. The stick is pivoted on
the torque tube for aileron control forces which are carried by
cable. The torque tube transmits the elevator control forces by
cable or push-pull tube. Wheel controls are generally mounted
on a small column which moves fore and aft for elevator control.
60
The airplane type of wheel coming through the dash panel has
not been used mainly because of installation difficulties.
The method of transmitting the control forces from the stick and
torque tube to the control surfaces varies in practice. In sailplanes
and utility wings a closed wire system is usually used. The aileron
cables end at the wing butt, on an idler horn which is connected
by a link to a bell crank on the fuselage. Thus the ailerons can be
disconnected by the removal of two pins. This eliminates the neces
sity of adjusting the ailerons when reassembling the ship. The same
thing can be accomplished by the use of a pull-push tube and bell
crank system. The fuselage idler horn is connected to the stick or
torque tube by a cable or push-pull tube system. Primaries com
monly use a closed wire system from the ailerons to the control
stick, or torque tube. While this system takes longer to set up and
rig, it is used because of its simplicity and low cost.
Elevators may be connected directly from the stick to the ele
vator horns by a single tube with proper guides to permit the use
of reasonable light tubes. Cables are more often used for the ele
vators than the tubes, however. The cables may be connected
directly to the elevator horns or to an idler horn and then to the
elevator horn by a link tube. When two separate elevators are used
they may be connected by separate cables or link tubes or by the
methods described before.
In all closed cable systems the linkage must be correct to avoid
loosening or tightening of the cable system due to unequal dis
placements of the horns. Quadrant type horns are superior to the
ordinary horn as they have a constant effective radius but are not
absolutely necessary except in special cases. They are used where
large angular displacements are necessary or if the direction of the
cable must be changed by a pulley close to a horn. The differential
action of aileron control is obtained by offsetting the idler horn on
the fuselage in favor of the up aileron. If extreme differential
action is desired this may be done by offsetting another horn: either
the horn at the stick or in the wing at the ailerons. In using differ
ential linkage it should be borne in mind that large forces can be
built up in the control system. The effect and magnitude of these
forces must be considered carefully.
As mentioned before, the arrangement of the control system to
facilitate rapid assembly is important. The use of push-pull tubes
and idler horns and link tubes is useful for this purpose and also
to eliminate the possibility of crossed controls. Where elevator or
ailerons are connected in setting up a glider it is advisable to use
61
some device to make sure that the cables cannot be crossed. Pins
of different sizes and different types of end connections for each
part will help prevent this from happening. A simple precaution
of this type is well worth while as serious accidents have been
caused by crossed controls.
Plain bearings for the control system are quite satisfactory if
reasonable care is taken with the design. If the glider is likely to
get extensive use, the use of removable bushings and other means
of eliminating play is desirable. Ball and roller bearings may be
used in all control joints and hinges to reduce friction and insure
long life. Grit and dust are a menace to plain bearings because it
is difficult to prevent their entry at the oil holes and other openings.
This grit will cause scratches that may be deep enough to affect
the strength of a thin walled tube. For this reason sleeves are rec
ommended for bearing points. Ball bearings are available with
dust shields and a permanent lubrication supply. This eliminates
the entry of grit and the necessity of frequent lubrications. The
cost of ball bearings is higher than plain bearings but it is not
excessive.
DETAIL DESIGN CONSIDERATIONS
One not familiar with glider and sailplane design might think
that once the general aerodynamic and structural design is decided
upon, the rest is of a more or less routine nature not requiring any
special design consideration. But such is not the case. The design
and arrangement of the details such as releases, controls, cockpits,
landing gears, etc., are special problems in themselves, requiring
much thought and consideration. The type and the purpose of the
sailplane are determining factors in the detail design, as are effi
ciency, safety, economy and simplicity. All these things must be
considered together with what is considered good practice in
designing these parts.
In European countries where soaring is more advanced, engineers
have developed special designs for parts that must be conformed to.
Releases, safety belts and fittings are some of the details that have
become standardized. This may tend to stifle inventiveness and
development, but it does result in safer and more uniform opera
tions. This idea could well be followed in the United States on a
more liberal scale.
Actually the whole design of a sailplane or glider can be broken
down into details, but we will be concerned with the more obvious
onesthose that are more open to change: releases, cockpit design,
62
extra control devices, landing gear, fitting and assembly features

and trailers.
Towing Releases and Hooks
A number of years ago when shock-cord launching was practically
the only type used, all gliders were equipped with nose hooks
which served as a means of launching and towing on the ground.
This hook of simple open type was mounted on the nose of the
glider. It was inclined backwards so that when there was tension
on the cord the ring would slide up into a rounded notch at the
top. When the tension of the cord was released, the ring slid down
and fell off. Other types of launches have almost completely re
placed the shock-cord method and most ships now do not even have
a nose hook. However, as there are still sites where shock-cord
launching is necessary, this method is not completely eliminated
from use. It is possible to use the release hook for this type of
launching, but there is some chance of trouble if the pilot does
not release at the right time.
In designing the nose hook, ruggedness and wear must be con
sidered along with the strength required. It must be rugged enough
to withstand rough usage and side loads encountered in launching
and ground towing. The minimum design load generally used is a
2400-pound pull forward, diverging as much as 14 in any direction
from a straight line through the nose hook and point of hold-back.
This is a considerable load and represents the maximum possible
with four strands of 5/g-inch. cord at 100% elongation (with a safety
factor). This is the tension design load for the hook, attachment
and usually for the front of the fuselage. The position of the hook
is important for proper operation. It should be placed where there
is no chance of the ring or cord catching in some part of the
fuselage when it drops off.
When auto towing replaced shock-cord launching, many types of
hooks and voluntary releases were designed. The ordinary nose
hook was unsatisfactory for towing as it was difficult and some
times impossible to release when desired and was generally unre
liable. The ring also had a tendency to fall off in flight and in
towing on the ground. The essential principle of a release is that
of a hook that can be opened at the will of the pilot. The impor
tance of proper operation can readily be seen, yet there were many
poor designs that failed to open or "jammed." In many European
countries the type of release has been standardized, but in the
United States no standard has been set and many types are being
used.
63
In Germany there is only one standard type of release allowed

for glider, tow plane or car. This is known as the DLV release and
is the result of extensive research and experience with many types
of releases. Its main feature is that it will operate regardless of
the position of the towrope, and that its opening does not depend
upon the tension of the rope. The release is really composed of two
parts: the two links and the release mechanism itself. The two oval
^THIS> PART MUST BE

LO SO THAT THE.
RING CANNOT JAM
AROUND IT
RING IN SELF
' /
RELEASING POSITION/ .
\
froR EMERGENCY
NORMALLY
\
TOWLINE RING IN NORMAL POSITION
'
L
BY PILOT)
SELF-RELEASING
TOWING
TYPE
RELEASE
RUGGED ATTACHMENT TO SUIT

TYPE Of CONSTRUCT/ONWOOD OR METAL. C.TC.
RING IN POSITION
UNDLR TENSION
LAUNCHING
SHOCK CORD
LAUNCHING
HOOK
TOWING RELEASE AND SHOCK-CORD HOOK

metal links are of different size, the larger being attached to the
towrope. The smaller one fits horizontally into the jaws of the
release and the larger rests against a metal ring welded to the
front of the release. This allows the rope to go in any position with
out any tendency to jam the jaws. The positive release instead of
the usual tension type insures the ejecting of the rings regardless
of whether there is any tension on the rope or not. So the only
possibility of failure would be to have the actuating mechanism
between the release and the pilot fail. As this is of simple and
straightforward design, such a possibility is remote.
64
One type very popular in this country, with an advantage over

other types, is the release that opens itself when there is any back
pull on the rope. This feature is very valuable when the towline
fails to release due to mechanical failure, jamming or pilot error.
The release is generally the same as other types except that the
hook and release trip are placed so that when the line of action of
the towrope is to the rear, the ring slides down the hook and
releases itself. This type does not have the positive release, but its
necessity is questioned by many. In almost all cases the tension
required to open the hook is exceedingly small and usually can
be accomplished by the weight of the ring and a few feet of the
towline. In the release above, this can be accomplished sometimes
by the weight of the hook itself. The one advantage that the DLV
release has over this type is that it will open from any angle of the
towline. Most self-releasing types will not open readily from direct
side pull but this condition is seldom if ever experienced and can
be corrected by changing the position of the ship.
There are many different types of releases being used. Many use
the principle of a pivoted hook with a catch that holds the hook
closed. When the release is pulled the catch opens and the tension
on the rope rotates the hook and throws out the ring. As the main
requirement for releases is that they function properly at the right
time for all conditions, these last three items should be considered.
The position of the release on the glider is important, as it can
affect the loads put on a ship to a large degree. The general design
makes it easy and convenient to put the release on the front or
underside of the nose, corresponding to the skid end of a primary
glider. This has proven to be the best place as it has the effect
of damping the action of the wings and tail surface. If the release
position is moved toward the center of gravity this damping effect
becomes less, towing becomes unstable and it is possible to put large
loads on the ship. Clearance of the rope and ring after releasing
is important. The fuselage around the release should be clear of
any obstructions that might catch the rope or ring.
As most releases will be used for shock-cord launching some
time or other, they should be designed for these loads. This tension
condition will more than cover those of airplane and auto tow and
usually designs the front of the fuselage for tension. Design loads
for winch and auto tow depend upon the general size and fea
tures of the ship and upon the towing speed. These loads should
be investigated thoroughly because they are surprisingly high and
side loads must be accounted for to insure ruggedness.
65
Cockpit
In cockpit design the main motive should be one of safety and
efficiency. It is the control point of the ship and houses the pilot.
In the past designers have not given much consideration to the
pilot's safety and comfort or to making the cockpit attractive. They
were concerned with improving performance and cutting down
the fuselage area and drag. Now the value of pilot efficiency and
comfort is realized and it is becoming the designing factor. The
effect of a comfortable upholstered cabin or cockpit on the skep
tical observer is also being realized. It gives a feeling of security
quite different from that given by seeing a pilot mixed in with a
mass of structural tubes or sitting out on the "front porch" of an
open primary.
The question of open or closed type of cockpit depends to a
large extent upon the type and purpose of the ship. Closed cabins
are used on sailplanes to improve performance and to protect the
pilot. They are not used on training ships because the feel and
sound of the air stream is one of the important aids to the be
ginner. In the closed type, flying has to be done mainly by feel
of the controls and by instruments. Many models are fitted with
convertible cowls so that both types can be used on one ship.
In all types of ships, visibility is important for safe flying. In
sailplanes, extra visibility is necessary for cloud and thermal soar
ing, especially in the upward quadrant, and the present trend is
toward larger and clearer cabins. Molded types of transparent en
closures give the best aerodynamic form but usually are expensive
and difficult to form. Flat celluloid sheets are most commonly used
and they result in a reasonably smooth section if small enough
segments are used. Cockpit enclosures should be made to open
quickly to permit leaving the ship rapidly in an emergency. The
attaching hooks or snaps should be of the positive type so that
there is no danger of their opening unless actually operated by
the pilot. Suitable ventilation control should be provided as an
enclosed cabin tends to get hot and stuffy. Colored top for trans
parent covers will help keep the glare down.
The cockpits and cabins should not contain any sharp corners
that one might bump against in a crack-up or a hard landing. All
edges should be suitably padded and covered over. The instrument
board should be smooth and padded where one might hit oneself.
Windshield edges and cowling edges should be taped with binding
or some similar material. Head and back pads should also be pro-
66
vided. A well-padded cockpit can do a lot to reduce injuries sus

tained in crack-ups.
On most gliders and intermediate sailplanes the stick control is
used almost exclusively, while in sailplanes, where small outside
dimensions are desired, the compact "dep" or wheel control is
widely used. Other "trick" types of controls have appeared but
even if they have many advantages they are not used much be
cause of the difficulty of getting pilots to change from one to the
other. Rudder pedals have almost completely replaced the rudder
bar, except in primaries, because of the ease of installation and
operation. There is little choice in the placement of these controls
except perhaps that the height of the rudder pedals is variable.
The design and shape of the seat is an important item for com
fort and efficient flying. People's tastes vary greatly as to the design
of a seat, but the fundamental thing is to get the weight distributed
so that no one part of the body carries too much weight. As para
chutes are worn for most soaring flights, their size and weight must
be considered. In training ships there is not much use for para
chutes as almost all the flights are under the minimum height
requirement for proper operation of the parachute. But in sail
planes it is the practice to wear one and it is required for airplane
towing, cloud flying and acrobatic flying. Provision is best made
for them in the seat. When one is not worn the well in the seat
can be filled up with cushions. The safety belt should be of the
approved type and fastened to the main structure of the fuselage
so that the full strength of the belt can be developed.
Release, spoiler, flap, brake and any other specialized controls
should be placed in convenient positions so that they may be op
erated easily and not confused. The release control is usually a
ring placed prominently on the dash. As we are accustomed to a
lever for the brake, it is best if this can be carried out for brakes on
gliders. Spoilers are sometimes coupled with the brake so that they
work together. It is necessary to hold the spoiler when first touch
ing the ground, for if it is closed the lift will build up again and
the glider may take off. So, in order to stop short, both the spoiler
and the brake must be operated at the same time while controlling
the ship. Flap controls must be of the irreversible type; that is,
designed so that loads on the flap cannot be carried back to the
control handle. This is to prevent a sudden gust or an increase in
load from taking the control out of the pilot's hand and closing
the flap. Other controls should be placed in accordance with their
importance and operating requirements.
As instruments are playing an increasingly important part in
67
soaring, the instrument boards are getting larger. Thermal soaring

and blind flying in clouds require almost steady instrument flying
and so the board in sailplanes must be placed where it can best be
seen without cutting down visibility. Other design considerations
of the cockpit also limit the position of the instrument panel. It
should be made so that it is easy to maintain and remove. In
sailplanes, provision should be made for extra equipment car
ried for long flights and contest and record attempts. A special
compartment with padding mounts should be provided for the
barograph. Room should be provided for maps, navigation equip
ment, radio, food, supplies, tools, camera, first-aid kit and other
equipment used on a long flight. Oxygen equipment is necessary to
exceed the altitude records.
Extra Control Devices
The modern sailplane has become so efficient that at times this
feature becomes a distinct disadvantage. With the high gliding
ratios and low sinking speeds it is a difficult problem to bring a
sailplane into a small field without the use of abnormal maneuvers
or special control devices. There are also times when it is desirable
to change the aerodynamic characteristics in flight in order to
increase or decrease efficiency, cruising speed, sinking speed or
gliding ratio. The use of extra controls for this purpose is just
beginning and it offers great possibilities for improving sailplane
performance.
As all these control devices involve aerodynamics to a large de
gree, they must be considered first from this angle. It is dangerous
to experiment without proper investigation as to the effect of the
controls upon the balance, strength, and aerodynamic features of
the sailplane. This is especially true when the controls are large
in size and hence may become powerful. Many of these devices
disturb the normal airflow, setting up disturbances that may induce
flutter in other components of the ship. Some tend to change the
balance of the ship and set up local loads that may prove trouble
some. These effects can be predicted and calculated quite closely
and so should be investigated thoroughly.
The need for devices to help sailplanes land is quite apparent.
If one comes in too high it must be slipped, fishtailed, or put
through some other maneuvers to use up its potential energy.
These are rather difficult in large-span sailplanes. With the use of
special devices these maneuvers can be eliminated and their effect
much improved and better controlled. One of the first attempts at
landing control was the use of split flaps. The effect of these flaps
68
is to increase the lift (decrease the landing speed) and increase the
drag with resulting decrease in gliding ratio. This was not quite
what was wanted for good landing control as the ships of that time
had very low landing speeds to begin with, and the lessened speed
resulted in sloppy controls and made them difficult to land in a
strong wind or rough air. However, with the rapid increase of
wing loadings and the use of flaps for variable airfoil effect, full
trailing edge flaps now have a definite place on sailplanes to de
crease landing speed and to improve performance.
In order to get better landing control, the spoiler was developed.
The duty of the spoiler is to decrease or "spoil" the lift and to
increase the drag, hence decreasing the gliding ratio and increasing
the landing speed. The spoilers are plates on top of each wing that
cause the lift to break down and the drag to increase over that
section. Their total area is usually around 1% of the wing area and
their aspect ratio from 4 to 8. They are more practical than flaps
because they offer better control with no tricky features. They can
be put on or off at will without any change of balance and any
danger of spilling. Because of this they are applicable for training
ships where the necessity for a quick landing often precedes the
pilot's ability to slip or maneuver the ship properly.
The optimum size, position and aspect ratio are controversial,
as not much research has been done along this line. They usually
are placed near the 30% point of the wing chord although the
present tendency seems to be to put them back a bit farther.
Their position along the span is of importance as they must be
placed so that turbulence from them does not affect the ailerons or
tail surfaces. It is best to have them in as far as possible so that if
one of them should become inoperative in flight, no great differ
ence in banking or turning moment would be experienced. As
they are usually placed where the pressure on the wing is lowest
they have a tendency to come open in flight. They should be de
signed so that this will not happen as it naturally has a bad effect
on the performance of the ship.
Another type of flap that has a place on sailplanes is the drag
flap. This flap does little or nothing to the lift but increases the
drag considerably. This type is also very adaptable to sailplanes
because of its simplicity and desired effect of steepening the glide
without appreciable increase in landing speed. This flap usually
is quite a bit larger than the spoilers and is placed on the underside
of the wing. Its position along the chord of the wing is not so
important as long as it is kept within limits.
The spoiler and drag flap have been combined in some cases
69
and called the double spoiler. The drag flap is placed directly
under the spoiler and may work with it or independently. The
effect is to decrease the lift and increase the drag considerably. This
type of spoiler was designed primarily for limiting the speed of
sailplanes in clouds. Due to blind flying and terrific turbulence,
excessive speed is sometimes gained rapidly. With the double spoiler
it is possible to increase the drag so that it limits the speed and
helps prevent failures. This type of spoiler is also very useful for
losing altitude quickly and to keep from being drawn up into a
storm front or a cloud.
A great future lies in the use of special devices to improve the
efficiency of sailplanes. The use of the adjustable stabilizer offers
one-piece elevator efficiency with hands-off flying qualities. The
fixed stabilizer is only efficient for one angle (or speed) and becomes
an extra source of drag at high arid low speeds. There is much to
be gained over the fixed type and the operation mechanism is rather
simple. Trimming tabs are fundamentally inefficient and should
be avoided on sailplanes. By deflecting full trailing-edge flaps a
bit upward it is possible to improve high speed performance, while
a slight downward deflection will slightly increase the lift and
efficiency. Further deflection will give normal flap action. These
features are especially desirable in cross-country sailplanes where
both minimum sinking speed and high cruising speed are desired.
The external airfoils offer good possibilities along this line. In
this type an auxiliary airfoil is placed a little behind the trailing
edge of the main wing. Its chord is usually from l/5 to % 0 of the
wing chord and it is supported by brackets coming from the main
wing. By varying the angle of the auxiliary airfoil the character
istics of the main wing are changed and reports show that this
method is much more efficient and effective than the full flap. It
offers some construction problems as it must be made accurately
and the wing must be strengthened to take care of the extra tor
sion. For the long-distance sailplane of the future, which will have
to fly fast and still have low sinking speed, it offers excellent pos
sibilities.
Landing Gears
The early gliders and sailplanes usually were equipped with a
simple elementary skid. Because of slow landing speed and light
weight it was possible to make smooth landings with this type. In
most installations the skid extends from the nose to a small dis
tance behind the center of gravity. Skids are usually made of hard
durable wood, sometimes covered with metal, and are easily re-
70
placeable. The "Vampyr" used soccer balls for its landing gear.
This was probably the first pneumatic type of landing gear used.
Instead of continuing along this line, engineers went back to
the skid, and used rubber blocks, tennis balls or springs for shock
absorption.
The skid type of landing gear was practical for shock-cord launchings, the extra friction of the skid making it easy to hold back the
tail. It also made possible short landings, as pressure could be
applied to the skid by nosing down. But with the advent of auto,
winch and airplane towing there was need for a better type of
landing gear. With these types of towing, the skid made it very
difficult to take off because of the high ground drag. It took a lot
of power to get these large sailplanes up to flying speed and towropes broke frequently and skids wore out quickly.
In Germany, instead of going to landing wheels, they developed
a two-wheeled cart called a "dolly" that fastened underneath the
skid and was used for taking off and handling on the ground. A
special fitting with a release was provided on the bottom of the
fuselage and as soon as the glider took off the pilot released the
dolly. This method, although reducing the drag, was not very
satisfactory in training and routine flying. It is a lot of trouble to
lift the ship and attach the dolly and there is always the danger that
a premature release might damage the rear part of the fuselage or
tail surfaces.
In the United States, wheels were introduced on training gliders
and it was not long before they were used on sailplanes. They are
much more convenient than a dolly and give better and easier
ground performance in handling and landing. With the use of
wheel brakes and the front skid, it is also possible to make very
short landings. Although the original cost is higher than the skid,
the cost of maintenance is very low. It has so many advantages and
good features that it is considered almost a necessity, especially
with the modern methods of towing.
Although both single and double type landing gears are used,
the single wheel partly enclosed in the center of the fuselage is the
more popular because it is simpler, cheaper and more efficient. In
most cases two-wheel landing gears are awkward and add a lot of
drag and weight. They have their place in training gliders where
they eliminate the man who holds the tip and consequently speed
up handling. In landing they also have the advantage of saving the
tips from damage. With the single wheel, tip skids are often used
to prevent damage of the tip but they are not absolutely necessary,
71
for if care is used the glider can usually be brought to a full stop
before the wing will drop.
There are many variations of the single-wheel type used. A prac
tical one is that with the wheel slightly behind the e.g. (center of
gravity) with a simple skid in the front and a spring skid at the
tail. In the normal loaded condition on the ground there is a little
weight on the front skid and the rest on the wheel. This makes it
convenient for handling and towing. In landing, pressure can easily
be put on the front skid by nosing down, if braking of this type is
desired. When the ship is empty it rests on the wheel and the tail
skid with most of the weight on the former. This prevents pound
ing of the tail in towing back empty and makes it generally easier
to handle.
Another variation of the single-wheel gear is the type with the
wheel well ahead of the e.g., with only a tail skid. The absence of
front skids makes the brakes very important as there is no skid to
rub along the ground. This type also puts a lot of load on the tail
which results in heavy handling and pounding in towing back.
Probably its only advantage is its simplicity of construction due
to the absence of skids and necessary supporting structure.
By moving the wheel position, the characteristics of the landing
gear can be changed. The farther back the wheel is moved (up to
the e.g.) the easier it is to handle on the ground, as the e.g. of the
ship empty is near the axle. The farther front the wheel is moved
the better it handles in towing, as the e.g. (loaded) is near the axle.
The disadvantage of the first is that when moved back too far
the loads on the front skid become too large. With the second, the
disadvantage is that the tail and handling loads become large.
Another type used often is that with the wheel just a bit ahead
of the e.g. (loaded), with a skid to the front and rear of the wheel.
In both conditions the weight rests on the wheel and the rear skid
with the tail skid off the ground. This type has the disadvantage
of rocking about the rear skid and banging the tail when towed
back empty.
Because of the high cost of self-contained wheel brakes, they have
not been used much in gliders. Most types have simple elementary
friction brakes that rub against the tire. A hinged paddle with a
canvas shoe is often used. A spring keeps it from the wheel when
not in use and a wire pull generally operates it.
The purpose and the type of ship should be considered in decid
ing upon the type and strength of landing gear. It is quite evident
that training gliders should have more rugged installations than
72
sailplanes. Also the question of drag is important in the higher

performance types and in some cases has led to retractable landing
gears. As there are usually no shock absorbing devices on gliders
other than the wheel, it is important that this be large enough to
take the required shock. The supporting structure should be very
substantial so that the wheel will fail before the structure will be
damaged. The effect of side loads on the supporting structure
should be investigated. A rugged supporting structure will do away
with a lot of repairing and realignment usually necessary after very
hard landings.
Fitting and Assembly Details

As fittings hold together the major components of a sailplane,
their importance can readily be appreciated. In designing and
constructing them many things have to be considered besides the
strength required to carry through the flight loads. As the wings
and other parts are usually handled by the fittings in assembly,
loading on trailer, etc., the fittings have to be rugged enough to
take these loads. They also have to have extra margins to take care
of wear caused by frequent assembly and disassembly. They have
to be protected against corrosion as they are often exposed to the
weather and scratched and worn off in handling. Plating is the
best method of protecting against corrosion, although proper
priming and painting will do if given good care. In some cases
fittings are used to support the glider on the trailer. This possibility
also must be considered in their design and allowances made for
strength and wear.
The materials used for fittings should be strong and tough, made
to aircraft specifications. Chrome-molybdenum steel is one of the
most popular metals for fittings, it being a tough steel that can be
welded easily. Materials which tend toward brittleness or softness
should not be used. Castings for fittings are taboo unless one has
equipment to guarantee their reliability and uniformity. Even then
the 100% margin required for castings usually overcomes any weight
advantage. The most common metal for pins is aircraft nickel-steel.
This can be machined easily in the heat-treated state.
Quick and easy assembly can do a lot to take the drudgery out
of setting up a ship. Of the three general types (cantilever, strutbraced and wire-braced semi-cantilever) the strut brace is usually
the easiest to assemble, with the cantilever next and the wire-braced
last.
In cantilever types it is the practice to use taper pins for assembly
as it is necessary to hold the wings rigidly in place. Because of
73
the long span and the closeness of fittings any play at the center
fittings will be magnified many times out at the tips. The taper
pins are pulled snugly into the holes by means of a locking nut.
To extract them the other end is usually provided with a thread
which permits the use of a puller. Cantilever wings take longer
than braced wings because the taper pins require more time for
proper fitting than straight pins. In a braced model the straight
pins are just pushed in and locked, while in a cantilever type the
taper pins have to be inserted, gradually tightened, and then locked.
\ VII iv \ vl vv \ \. V \ \
THREAD fo
PULLER NUT
TYPICAL CANTILEVER SPAR JOINT DETAILS

In semi-cantilever design with struts it is conventional to use
straight pins for fittings. Although tightness of joints is not so im
portant as in cantilever types, it is important to have good, smooth
fits. If there is a little play to start with it will quickly wear the
hole larger. Regular aircraft nickel-steel bolts are often used for
pins but their manufacturing tolerances sometimes result in loose
fits. On strut ends, universal points are widely used to prevent
damage to the fittings. They prevent side loads from being carried
through the strut into the fitting.
Although putting the wings on may seem to be the main and
longest job in assembly, it usually takes more time to put on the
74
tail surfaces and hook up the various controls. Because of this,

extra time spent on simplifying assembly features will be amply
repaid in ease and speed of assembly. With some types it is possible
to leave the horizontal tail surfaces on in trailing and with many
the rudder can also stay on. However, the majority of sailplanes
have to remove their tail surfaces for trailing. Folding tail surfaces
are quickly coming into vogue as they simplify assembly and carry
ing of the surfaces. There are, however, some design and construc
tion problems that sometimes make this type impractical. The
substitution of push rods for wires or cables will also help to speed
assembly, and prevent incorrect hooking up. Assembly methods and
systems offer a field for clever ideas and invention, and effort along
this line will not be wasted.
Trailers
The main requirement for a glider trailer is that it transport the
glider or sailplane from point to point without damage. A ship
being transported is liable to be damaged due to: improper loading
or suspension, wind, rain, hail, and the numerous accidents of the
road. How far one wants to go to protect his craft depends a lot
upon its type and cost. However, there are fundamental require
ments that should be observed so that no structural damage is
done to the glider.
The trailer should be sprung properly so that shock and vibration
will not disarrange the mountings or damage the parts. The parts
should not be mounted so that large loads can be put through the
glider structure. All parts should be held firmly in place so that
no damage results from the moving or wearing of parts. The trailer
must operate properly without trouble from running gear, struc
ture, hitches or lighting equipment.
For primary and utility types the open frame type trailer is
widely used. It is simple and inexpensive and is satisfactory for
these types of ships. For more expensive types the covered trailers
are more popular because of the added protection that they give.
They are particularly desirable for wooden ships where protection
against weather is important. They naturally cost more to build
but they are really worth while and simplify the storage problem.
A well constructed box trailer serves as an excellent storage place
for sailplanes.
There are many types of closed trailer in use. The simplest type
uses a canvas cover over the open type trailer, it being held in
place by a few posts and tie-down lugs. The simple open type can
be converted into a canvas box type with the addition of the neces-
75
sary framing members and canvas covering. The most durable, but
heaviest, type is the wooden box where the whole trailer is com
pletely enclosed with a covered body. A variation of this is the
type that has a permanent top but removable canvas sides that
permit the wings to be loaded easily. In the completely covered
type the wings and fuselage have to be put in from the end, which
sometimes is a lengthy and tricky job. The same protection with
easier loading features is the advantage of the removable shell type.
These shells of light metal framework and fabric covering fit over
the trailer and completely enclose it, fastening by hooks.
The simplest and most widely used type of trailer running gear
is the two-wheel type with the wheels somewhere near the center of
TRAILER FOR MINIMOA

Left wing has been removed
gravity. Four-wheel trailers have been used but they are not very
practical, costing more for original parts and upkeep with twice
as much chance for trouble. In most cases the front axle and wheels
of a light automobile work best for the running gear. If the total
load is to be light, or if a heavier type of front end is to be used,
some of the leaves of the springs should be removed so that it
will not be too stiff. Trailers have been built with no springing
except that provided by the tires, but this type is very rough on the
glider.
The chassis or main framework of the trailer has been made in
many types and materials. The most common are the rectangular
or triangular in plan form constructed of wood or steel tubing.
A good grade of lumber and low priced commercial tubing are
suitable for trailers. The wooden type is easiest to construct for the
beginner but the steel type results in a light, efficient trailer that
will last a long time and be easily adaptable to other requirements.
The main feature of the chassis should be strength, to carry
76
through the necessary loads and stiffness so that deflection of the

trailer will not strain the glider. The draw bar to which the trailer
hitch is attached should be securely held in place, for failure of the
draw bar has resulted in many a nailer breaking away. It is an
advantage to have the chassis outline enclose all parts of the
glider so that it can act as bumper. If the wings are just hung on
outside the frame they are very liable to damage.
The mounting of the ship on the trailer is very important. The
various parts should be mounted so that no large loads will be
carried through the structure due to the deflection of the trailer
or to the method of suspension. In this respect it is good practice to
avoid putting any loads on the structure; rather let the parts lie in
place, secured by properly padded supports. All movable surfaces
should be clamped together and padded out where there is any
chance of wearing or chaffing. Generous use of padding will help a
lot to prevent minor damage. All clamps and pins should be of the
locking variety so that vibration over a long period will not loosen
some important clamp.
It is important to keep the trailer reasonably light so that it will
not burden the tow car. Also, the weight carried on the tow bar
should not be large as it burdens the springs of the car and makes
the trailer heavy to move by hand. However, if the load is too light
or if the trailer is unbalanced, the hitch has a tendency to rattle and
wear and uncouple itself. There are various types of hitches avail
able but only those of sturdy design with locking devices should
be used. Safety links between trailer and car, required by many
states, should be installed. The trailer lighting should comply with
the state regulations and be sufficient to prevent accidents from
improper lighting. The use of turn signals is advisable.
For those who are unable to build their own ships, the trailer
offers an excellent opportunity to exercise their ingenuity and in
ventiveness. There is practically no end to the variety of trailers
possible and there are no strict regulations that limit one's work.
Extra time spent on trailers is well repaid in trouble-free, worryfree trailing.
MAINTENANCE
Proper maintenance of equipment is an important requirement
for trouble-free gliding and soaring. Here, equipment is not meant
to include only gliders and sailplanes, but also instruments, towropes, trailers, tow cars, etc. Of course, failure properly to main
tain the flying equipment will result more likely in trouble than in
neglect of any of the others, but they are also very important. Too
77
much faith in an instrument or in the strength of the towrope can

sometimes cause as much trouble as overloaded damage to the ship.
Frequent inspections and maintenance checks take little time
and are a good insurance against accidents from that source.
Because of the essential lightness of gliders and sailplanes, they
naturally must be treated with more care than airplanes. Their
lighter construction is more susceptible to damage than the more
rugged structure of the airplane. Of course, this varies with the
type of ship, the primaries and utilities usually being more rugged
than the sailplanes. As a matter of habit and good principle, all
types should be handled carefully. Lifting in the wrong places,
dropping and bumping are some of the main faults of improper
handling.
In assembly a definite routine should be adopted to speed it up
and avoid mistakes and possible damage. In most cases it is best to
take the parts from the trailer as they are needed. Having the
various parts spread all around should be avoided as they are in
danger of being stepped on, run over or blown away by a sudden
gust of wind. A brief inspection of parts made inaccessible by as
sembly should be made before assembly is started. If fitting pins
are not put back in place after the ship is taken down, they should
be marked so that the same pin is used each time. This is especially
important with taper pins because a very good fit is necessary for
ease in assembly.
In assembly, moderation should be the keynote. If something
does not fit do not use force or "the hammer" until you have in
vestigated to see what is causing the difficulty. The part may have
been damaged in handling or trailing, or some foreign substance
might be making the tight fit. The controls should be hooked up
last, for if they are hooked up too soon they may be strained by
the movement of some semi-attached part.
Once set up the ship should be thoroughly inspected and con
ditioned for flying. The inspection is best carried out with the aid
of "fill-in" cards that list the various parts to be checked. This will
tend to do away with the possibility of overlooked parts in the mad
rush to get the ship in the air. Conditioning for flight usually in
cludes: checking inflating of tire, oiling of moving parts, wiping
wings and fuselage, and adjusting the various instruments, parts
and controls.
In disassembly the same care should be taken as with assembly.
The controls should be unhooked first and pins put back in place
as soon as possible to prevent losing them. If the ship has received
any rough handling in the air or in landing it should be inspected
78
for possible damage before putting back on the trailer. In loading

back on the trailer, the troughs and pads should be inspected so
that any foreign substances lodged there can be removed. Dust and
dirt tend to act as grinding compounds and wear the finish off.
Extensive inspections should be carried out every so often, de
pending upon the extent and character of operations. In this
inspection every part that it is possible to see without opening any
fabric or plywood should be thoroughly looked over for wear, de
terioration and damage. When the ship is showing signs of wear
and the fabric is getting saggy and porous, it should be recovered
and completely gone over. All worn parts should be repaired or
replaced and the whole structure given a protective coat of varnish
or the like. In metal structures this procedure is considerably sim
plified. For minor repairs, of all types of constructions, parts can
be replaced or repaired. But with major repairs the drawings and
requirements are usually necessary to insure the use of proper
methods and materials. Accepted methods and aircraft materials
should be used for all repair work.
The maintenance of the trailer is an important item, for trailer
trouble can easily spoil a soaring day. The various supporting
brackets should be inspected for damage and wear and misplaced
pads. The running gear should be inspected for mechanical trouble
and lubrication, and the tires checked for wear and inflation. The
hitch, chassis and lighting system should be looked over to make
sure that they are in good working order. The structure should
receive protective paint when necessary and the canvas covers should
be treated for watertightness.
Storage, although a simple problem for metal ships, is not so
simple for wooden ships, for too dry or humid conditions of storage
tend to weaken the glued structure. A place of average humidity
and temperature is best. Gliders are easily stored in their trailers
if their mountings and paddings are designed properly. However, if
wings are stored out in racks, they should be placed so that there
is no strain on the wings or large local loads on any small portion.
One method of storing is to leave the glider assembled, but this
requires a great deal of room which is not usually available. If the
ship is stored for a long period it should be inspected occasionally
to make sure that everything is in good order. Leaky roofs, con
densation and mice sometimes are the causes of unexpected trouble
in storage.
Properly maintained equipment considerably reduces the possi
bility of trouble. It is a comforting thing to know that your ship
79
is in good order and ready to take its full design loads. These extra
inspections do not take long and usually they can be done when
things are slow: while waiting for the tow car to come or for a wind
to pick up.
THE OLYMPIC SAILPLANE

According to the rules of the 1940 Olympics, planned for Finland,
all pilots competing in the contest must fly the same type of sail
plane. The design of this ship was selected by the F.A.I. General
specifications were issued but were of such a nature that designers
were allowed a great deal of latitude. Out of five ships from Ger
many, Italy, and Poland, the D.F.S. "Meise," of Germany, was
chosen.
The most outstanding feature of the D.F.S. "Meise" is the fact
80
that, although the performance is excellent, the main purpose of the

design is to provide a ship that can be built and handled by inex
perienced workmen, without the use of expensive or complicated
tools. It is a high wing type, of standard plywood construction with
semi-monocoque fuselage and full cantilever wings and tail. There
are no complicated welded fittings or parts. It can be assembled by
three men in eight minutes, and disassembled in four minutes. The
specifications are as follows:
....... . . ........ . .... .49.5 ft.
Span ..............
Wing area ..... .. ................... .... 161.0 sq. ft.
Aspect ratio ....... .................... ............... 15.0
Wing loading ............ . ......... ......... 3.09 Ibs./sq. ft.
Empty weight ..... ................................. 354.0 Ibs.
Gross weight .............. .........................496.0 Ibs.
Minimum sinking speed .......................... 2.2 ft. per sec.
Best gliding ratio ..... ................................ 25 to i
Stalling speed .................................... .31.5 m.p.h.
3.3
6.6
9.9
13.
DFS - Meife
r
31.2 M.P.H.
62.-)
93.8
Courtesy of SOARING
FIG. 2. PERFORMANCE CURVE
\\'i}l<r
The engineering of the wing was based on the wings of the D.F.S.
"Reiher" and the D.F.S. "Weihe." A straight taper of 2.6 to i is
used with an average chord of 3.3 feet. There is no gull, since it
has been found by experiment that the proper combination of
dihedral and rudder give sufficient stability on spiraling. The wing
section varies from go. 549 at the root (16% thickness) to go. 676
at the tip. The 549 section extends to (3o% of the semi-span. The
combination of the high CL of the tip section, and a seven degree
washout at the tip, insures excellent control at the stall. A dihedral
of 2.5 degrees to the neutral axis is used.
81
The wing is composed of a single D-spar with an I-beam web,

and a very light rear spar, which carries the aileron. The aileron is
hinged in four places.
The main root fittings are composed of four straps on each main
spar. To simplify construction, all of them are identical. Referring
y
Courtesy of SOARING
FIG. 3. MAIN ROOT FITTINGS
to Fig. 3, the wing is attached to the fuselage by pins A, and cor

responding pins in the rear spar. When pins B are removed, the
wings may pivot about pins A, so that both wing tips may rest on
the ground at the same time. This was done so that two people
Courtesy of SOARING
FIG. 4. SPOILER MECHANISM
could assemble it easily. Both the wing and fuselage root fittings
are attached to the structure by tubular rivets.
The spoilers are shown in Fig. 4. This type of mechanism was
chosen both because of its low cost and because of the ease with
82
which it may be fitted to the wing contour. It is only necessary to

make them too big and then plane off the excess material. They
will not affect the wing contour by warping. It will be noted that
they move in a plane parallel to the spar, and do not rotate, as is
common in this country. They are enclosed in a plywood box that
keeps water and dampness from entering the wing.
Fuselage
The fuselage is reinforced by bulkheads composed of two cap
strips with a plywood web. They are so designed that only small
strips of plywood are required, thus making it unnecessary to use
Courtesy of SOARING
FIG. F;. PULLEY INSTALLATION

up a whole sheet for each bulkhead. The cross-section behind the
wing is almond shaped, for simplicity in covering and for maneuver
ability. For purposes of stability and sensitivity of control, the
fuselage is very long (24 feet).
Following standard German practice, there is no wheel. The
hardwood main skid is mounted on doughnut-shaped rubber shock
absorbers, and the spring for the tail skid consists of two tennis
balls.
There is a luggage and barograph compartment behind the main
bulkhead, which is accessible from the outside.
83
Tail Surfaces
The elevator and fin are of two-spar construction with stressed
skin leading edge. The spar and rib construction is identical, being
a plywood web with cap strips on one side only, to form a channel
section. The elevator and rudder are statically balanced, with a
torsion-resistant spar and straight ribs. There is a trim tab on the
elevator that can be operated in flight. The elevator and stabilizer
assembly is attached to the fuselage by one bolt and wing nut.
Control System
Only four ball bearings are used in the entire control system. All
other important bearings are bronze bushed and pressure lubricated.
There are no press fits, and only one size of reamer is used. The
control stick and torque tube installation are mounted on universal
bearings so that no alignment is necessary for installation.
All the pulleys are mounted as in Fig. 5, so that only one type
of pulley and pulley bracket is necessary on the entire ship.
Either of two simple rudder pedal installations is optional. One
is adjustable in flight. The other is not adjustable at any time.
The sailplane may either be built from plans furnished by the
factory, or it may be bought completely built.
CHAPTER V
LAUNCHING METHODS
I HE FIRST METHOD USED to launch a glider into the air

was that of Lilienthal and other early pioneers who simply ran
downhill into the wind until the light hang glider supported on
their shoulders became air borne and in turn supported them.
Since those early days a multitude of different methods has been
used to launch motorless aircraft. These include: releasing from a
hot-air balloon, release from a dirigible, towing behind galloping
horses, towing behind an automobile, launching at the crest of a
steep hill into a strong wind by having the glider pulled forward
by a man at each wing, towing behind a motor boat (seaplane
glider), shock-cord catapult, winch towing, and airplane towing.
Although the original method has been revived by an enthusiast
in California who has built a modern version of a hang glider, and
others are still occasionally used, only four of these methods have
been recognized as having practical value and are now in regular
use throughout the world. These, in order of their importance as
well as probable use by students are: automobile towing, winch
towing, shock-cord catapult, and airplane towing.
AUTO TOW
Automobile towing, proven to be the most practical and safest
method for student instruction, is also useful for launching for
soaring flights on large fields or on the top of ridges where there
is sufficient room and a winch is not available. Equipment needed
includes the automobile, release mechanism, towrope and metal
rings.
The choice of a proper tow car is important; some care should
be taken in acquiring one. The ideal type is a light but strongly
built roadster or touring car with plenty of reserve power. The
85 h.p. Ford V-8 is an example of an excellent tow car, although
many of the earlier Model A's have proven quite satisfactory and
have been used extensively.
The top of the car should be down or, preferably, removed alto-
LAUNCHING METHODS
85
gether so that the driver has an unobstructed view of the glider at

all times. It is also best to remove the windshield, both to eliminate
a possible hazard in case of accident and to allow the driver to
acquire a sensitive feeling of air speed on his face which makes
it unnecessary for him to watch the speedometer frequently. Al
though some tow cars have been rebuilt with one scat facing back
ward for the instructor beside the forward seat for the driver, the
best method is for these two jobs to be done by one man. If the
instructor is driving he can accelerate or stop the car in case of
emergency more quickly than would be possible if he passed the
order on to another. Also it does not take long to become so
familiar with the car and the field that most of the driving can
be done facing backward watching the glider.
It is sometimes advisable to have the rear of the car weighted
down with 200 or 300 pounds of cast-iron weights or flat boiler
plate, well secured to assure proper traction on rough ground. This
can also be helped by softening up the rear springs. It is a good
idea to use oversize tires at comparatively low pressure to prevent
cutting up the ground if operating on a grass-covered field. If
operating on soft ground or sugary sand such as is found on the
Michigan beaches it is necessary to use the super-balloon tires such
as were manufactured a few years ago. To get good traction even
with these tires, frequently it is necessary to keep the air pressure
very low.
The rear of the car must be equipped with an approved type of
release either securely bolted or welded onto the framework of
the car, preferably at least 2 feet off the ground. The chief purpose
of this release is to enable the tow car driver immediately to detach
the towrope if the release mechanism of the glider jams at the top
of a tow. To operate it a light rope of about 14-inch diameter
should lead from the trip of the release to a position within easy
reach of the driver. It is usually brought over the top of the seats
or around the left side of the car so that he can pull it with his
left hand as he drives with his right while looking back over his
left shoulder.
For primary training a i5o-foot manila towrope of z/8 - or even
y8-inch diameter should be used. Although this size of rope is too
thick for greater lengths and consequently higher tows due to its
high aerodynamic drag, it is better for this preliminary stage where
it is subjected to much dragging on the ground which will quickly
wear out lighter sizes. Five-sixteenth-diameter rope will do if the
other is not available, but 14-inch should not be used as it is only
just strong enough and will wear out very quickly.
86
Two other lengths of rope also should be on hand as the student

progresses; an intermediate rope of 300 feet of % G -inch, and a
5Oo-foot length of % G -inch rope for high tows. The intermediate
rope can be used for perfecting 180 turns on a field or launching
off the top of a ridge. The 5oo-foot length is used to enable the pilot
to climb to a maximum of approximately 425 feet to make 360
turns. On windy days or on large fields the two ropes can be joined
to make an 8oo-foot line enabling the pilot to get high enough to
make a much longer glide and, if the conditions are right, to en
counter thermals. When such lengths are used, flags or streamers
of colored rags should be tied at intervals along the rope to make
it visible to other aircraft.
CORRECT ATTACHMENT OF RING TO END OF ROPE TOWLINE

At both ends of each of the three towlines should be fastened
welded steel rings at least 2 inches in diameter and \/ inch in
cross-section for the average type of open releases, or the double
oval links for the DLV releases. These should be spliced on with
thimbles to prevent cutting of the rope and possible tangling of
knot ends. The splices also should be tightly wrapped with friction
tape. It is best to have the same type of release in the tow car as
on the glider to prevent possible delays in having to switch opposite
ends of the rope.
A valuable scheme to speed up high tows for 360 turns, when
only the tow car driver and glider pilot are present, was developed
by the author. This scheme is as follows: have four 2-inch rings
on one 5OO-foot length of rope. One is fastened at each end and
the other two are fastened 40 feet in from the ends of the rope so
that these rings become links in the rope which is fastened to
them by a thimble, eye splice and tape on each side.
The tow is made as usual with one end ring in the glider release
and the other in the car release. After the glider lands the tow car
is driven to it, the driver gets out, detaches the end ring from the
car release and attaches it to the glider release. He then pulls in
LAUNCHING METHODS
87
40 feet of the rope and puts the second ring in the car release.
With this short length he then tows the glider back to the starting
point fast enough for the pilot to maintain lateral control, the
remainder of the rope dragging behind. As the starting point is
reached he slows down and maneuvers the car to turn the glider
slowly around into take-off position, the down wing sliding back
ward on a pivoting, shock-absorbing wing skid. The driver then
pulls his release, dropping the 4o-foot ring, and drives off upwind
along the remainder of the rope which has been automatically laid
out in line by the tow back. He then jumps out, inserts the end
ring in the car release, climbs back in, starts the car slowly to take
up the 40 feet of slack and begins another tow. This procedure,
during which the glider pilot never leaves his cockpit between
flights, and the tow car driver jumps out only twice, enables a
flight to 400 feet altitude to be made every 5 or 6 minutes with the
glider being in the air nearly half of that time. As noted, only
three of the rings actually are used on one tow back, the fourth is
put in for convenience so that either end of the rope can be used.
Knots should never be allowed anywhere in a towline except,
perhaps, temporarily to save time. Wear from pulling the rope
along the ground will take effect much faster on a knot and will
soon result in a break. All breaks should be spliced together. In
making a short splice, unlay the ends of the ropes to be spliced
together. The ends are crotched. In splicing a rope of moderate size
the first tuck can be made by hitching together the opposing strands
in the crotch, as in the first part of a reef knot. Then taper on each
side. This gives one full tuck, and two tapered tucks on each side
of the middle tuck.
Cast or malleable iron rings should never be used as they may
crack when dropped and later fail under towing loads. Their origi
nal strength is also questionable.
Stranded metal cable and hard wire have certain advantages fol
lowing for students that have progressed at least to the stage of
making 360 turns but there are also several dangers connected with
their use. The advantages are very low cost as compared to
manila rope, rather long life when properly handled, and minimum
aerodynamic drag. Numbers 14 and 16 soft grade spring wire are
best suited for glider towing. Music wire of .O56-inch diameter has
proven satisfactory for lightweight gliders with a gross weight of
less than 500 pounds. A bad feature of wire is kinking. Every loop
may fold into a kink when the towing pull is again exerted and
during the hard pull of a steep tow the wire will break. A watch
should be kept for these loops and kinks by frequent inspection

of the line. When they are found they should be straightened out,
or if too sharp to be straightened, a cut should be made and the
ends spliced together. The use of a parachute of 2- to g-l'eel diameter
fastened near the glider end of the wire is essential and avoids the
kinks to a large extent. An old sock fastened onto the nose of (he
glider makes a satisfactory case for the folded parachute which is
pulled out by the weight of the rope after the release. When work
ing on hard wire with pliers or other tools care must be taken to
prevent injury to the surface of the wire as this may cause a failure
under load.
Tied fo prevent"
/ further
When starting
*4fc
, .
the strands of each
a/fernafe
From THE SEA SCOUT MANUAL. Courtesy of the

Boy Scouts of America
SHORT SPLICE
There is little or no resiliency to wire or cable so all the bumpings of the tow car on uneven ground are carried directly to the
glider, instead of being largely absorbed by the towline as is the
case when rope is used. This is apt to cause the structure of the
front part of the fuselage and the wings of the glider to be unnec
essarily and perhaps dangerously stressed. To prevent this a shock
link is used. The simplest and cheapest consists of about 25 feet of
i/2- or 3^-inch rope looped and fastened onto the car end of the
wire. This tends to lie out flat and therefore transmits nothing but
smooth pull from the tow car to the line. It serves also as a drag
when towing the line back on the ground and so tends to keep it
LAUNCHING METHODS
89
straight. It should be formed into a loop of two strands with a ring

to fit the car release. Also useful is a heavy spring of tightly coiled
i^-inch spring wire 2 inches in diameter and 18 inches to 2 feet
long. Another good shock absorber can be made of a 5^-inch shock
cord. The cord should be served into an endless loop of two turns
with \vebb straps riveted on, and provided with a ring on one end
and a harness snap on the other. A disadvantage of this is that it
soon will wear out from dragging over the ground. The spring and
shock-cord links should both have limit cables to prevent over
loading. These should allow from 50 to 75 per cent stretch of the
elastic part of the link.
All towlines, both rope and wire with the exception of the short
lines used for primary training, of greater strength than i^-inch
manila rope should have a weak link provided at the glider end.
This should have about the same strength as a i/^-inch manila rope
and may be an 8- or lo-foot piece of that material. The purpose of
this is to have the rope break before the glider becomes dangerously
overstressed.
The greatest danger in the use of wire for towing is with static
electricity. The glider traveling through the air acts as an excellent
static accumulator, especially when constructed with metal and fabric.
This may build up dangerous potentials if not conducted off to
the ground continually. A continuous electrical conductor should
be provided to the tow car from the glider and a metal drag chain
provided from the metal structure of the car to the ground like those
used on gasoline trucks for the same purpose. All gliding opera
tions where wire or wet rope is used for towing should be suspended
in cases of thunderstorms or any indications of other atmospheric
conditions with heavy static nearby.
Towlines and broken ends of rope and wire are dangerous when
left lying around on the field or airport. They may catch in parts of
a glider or airplane and cause an accident. Towlines not in use
for even a short timeshould be dragged to the edge of the field
out of the way, or, better still, wound up on a simple drum turned
by a hand crank.
The tow car should be operated only by an experienced driver.
If gliding instructions are being given he must also be an expe
rienced pilot. The best procedure with a powerful car pulling the
average glider is to start in second gear and keep in the same gear
throughout the tow. This should hold true for either a short instruc
tional tow or a high tow. The chief value of staying in this gear is
that the driver has better control over the towing speed due to
quick acceleration or deceleration thus possible. First or low gear
90
may be necessary to give sufficiently rapid initial acceleration to a

heavy sailplane. Third or high gear may also be used on a high tow
into a good breeze after the glider has reached 200 or 300 feet.
If any gear changings are made while towing the glider, they must
be made quickly and smoothly so as to give as little jerk as possible
to the glider. To allow for the possibility of the student pulling
back too steeply and overstressing the wings to the danger point
the tow car driver must be careful not to tow too fast. It is a safe
rule to limit this speed to 40 m.p.h. in still air and proportionately
less if a wind is blowing.
To give quick acceleration on ridge top launching fields of lim
ited area as well as to save wear and tear on tow cars running over
rough ground, a system using a pulley has been developed to a
high degree of efficiency in California. First used at the Torrey Pines
Tot* Car
AUTO-PULLEY LAUNCHING
Mesa site north of San Diego this system uses one pulley of about 12
inches diameter with oversize flanges attached to the car by a vertical
hinge on rear bumper. One end of a 3/^-inch rope is attached to a
low stake driven firmly into the ground. It is then brought around
the pulley and the other end, equipped with a ring, is attached
to the glider release. The car is driven in low gear into the wind at
half the flying speed of the glider. The speed of the car must be
reduced as glider nears top of climb.
To avoid risk of failure of the glider release mechanism, someone
with a knife or, preferably, a pair of sharp shears could stand near
the stake. Otherwise a cut-off mechanism should be built onto the
pulley.
In starting a tow the driver must be careful to drive the car very
slowly until the slack has been taken up entirely before using full
power to accelerate as quickly as possible. When a long rope is used
and a third person is at hand to hold the wing tip it is customary
for him to leave the wing tip down until the slack is out; then, to
show that the slack is out and also that the glider pilot is ready to
LAUNCHING METHODS
91
take off, he holds the wing level. When no third person is present
the pilot can kick his rudder full from one side to the other as a
signal to the driver.
The tow car driver must realize that his is a serious responsibility.
He must be keen and alert at all times. Before a tow he must make
sure that the car is filled with gas, oil and water, the tires held at
proper pressure, the steering mechanism tight and oiled, and the
engine running smoothly and warmed up. He must also make sure
that the field and air are clear. During the tow he must be con
stantly on the alert to prevent towing too fast or too slowly and
must always be ready to stop the car or release the towline in case
of an emergency.
THE EXPLOSIVE RELEASE
A safety release in order to be useful to the gliding public must be
simple and inexpensive to construct and, above all, must be fool
proof and under the direct control of the instructor. Since none of
the mechanical devices possesses all of these features, a rope-cutting
technique is suggested which offers these and several other features
LEAD WIRES
EXPLOSIVE
CHARGE
^SULFUR
ÂSPHALT
/
/
^ ASPHALT 8 SULFUR
/
^ FULMINATE OF MERCURY
CROSS-SECTION OF TYPICAL ELECTRIC BLASTING CAP

desirable in a safety release. In its final form, determined after a
series of experiments, the release consists essentially of an electric
blasting cap. When the small explosive charge in the cap is de
tonated with an electric current, the high-velocity gases from the
explosion shatter the rope. The following features recommend this
release to winch and auto-pulley operations:
a. Simplicity of construction.
b. Reliability achieved by the lack of moving parts,
c. Ease and low cost of reloading.
d. The report from the explosion is loud enough to be heard
by the pilot, warning him that the towrope has been cut.
Electric blasting caps are made in two sizes, the smaller of which,
designated as No. 6, is recommended for glider releases. A crosssection of a typical electric blasting cap is shown in the accompany
ing illustration. Although they contain a very small quantity of
92
explosive, blasting caps are, nevertheless, dangerous in the hands

of amateurs. For this reason a few rules for the safe handling of
electric blasting caps are suggested:
a. Store caps in a cool, dry place, preferably under lock and key
to prevent tampering by children or the inexperienced.
b. Do not subject caps to severe shock or extreme temperatures.
c. Do not carry caps loosely in pockets or in tool boxes where
they may be crushed.
d. Do not handle caps during an electric storm.
e. Do not explode caps near a crowd of people.
f. Do not expose caps to direct rays of sun.
g. Do not dissect caps.
h. Do not pull on the lead wires. This may break the bridge
wire, making the cap inoperative.
Since some states have licensing regulations for the purchase and
use of explosive, it is recommended that these regulations be
consulted before using blasting caps in those states. Generally, these
regulations permit the private use of small quantities of explosives.
It should also be mentioned that the transportation of explosives on
common carriers is prohibited by Federal law. It is therefore best
to purchase the caps in the neighbood of the flying field.
The explosive release is made from a block of mild steel, al
though tool steel is preferable if the release is to be used during
pilot instruction. The dimensions of this block are shown in the
accompanying diagram. The electric blasting cap is laid into the
groove and a wide rubber band is passed around the block and
over the cap to hold the cap in place. The release is installed with
the open side of the groove facing the towrope.
For use in the auto-pulley launching method the grooved steel
block is mounted on the pulley frame at a distance from the pulley
such that the distance from the cap to the rope does not exceed 3/4
inch (see diagram). This distance cannot be exceeded without af
fecting the reliability of the release. Pulleys of cast materials will
not stand the shock of the explosion and for this reason forged
steel or welded steel pulleys are recommended. The open end of
the grooved block should be pointed downward so that in the event
of a misfire the explosion will be directed away from anyone work
ing over the pulley. The release should be so mounted as to elimi
nate the possibility of a splice in the towrope hitting the cap and
dislodging it or even exploding it.
When used with a winch the explosive release should be mounted
on the level winder at a place where the rope cannot whip too far
away from the release. The explosive release should be placed near
LAUNCHING METHODS
93
the roller in order that the distance from the blasting cap to the
towrope shall not exceed s/4 inch. The release should direct the
explosive blast away from the winch operator and toward the
ground.
The current for firing the blasting cap is obtained preferably from
the storage battery of the tow car. An electric plug should be in
stalled to disconnect completely the firing circuit from the battery
except during the actual towing. This is a safety procedure which
8LAITIMC CAP
HOLDER ~
BLASTING
CAP HOLDER
THE EXPLOSIVE RELEASE FOR AUTO-PULLI v LAUNCHING

cannot be overemphasized. For firing the blasting cap a nonlocking
push-button switch should be placed on the ungrounded side of the
circuit. This switch should be mounted in the tow car on the dash
or on the steering column so that it will be readily accessible in an
emergency. The wiring should be done with flexible two-conductor
rubber cable. Special precautions should be observed to insulate the
circuit, especially the cap lead wires, from the frame of the car
and pulley or winch.
AN AID TO AUTO-PULLEY LAUNCHING
In taking-off for thermal soaring the probability of a successful
contact with an upcurrent increases when the sailplane is launched
to greater altitudes. For this reason a premium is set on a maxi
mum altitude being gained by launchings from fields of limited
size. In auto-pulley launching the maximum altitude depends not
only upon the maximum safe tow speed of the glider but also upon
the position of the stake anchoring the stationary end of the tow-
94
rope. If the stake is placed too near the glider at the start of the
tow, the altitude is limited by the rope length between the glider
and the pulley when the tow car has completed its run. On the
other hand, if the stake is placed too near the end of the field
opposite the glider, the launching run available to the tow* car sets
a limit to the maximum altitude. The problem in the auto-pulley
method is that of determining the stake position which permits
a maximum altitude to be reached from a given field. In the past,
several empirical rules have guided auto-pulley operations from
small fields.
The mathematical solution to the above problem for any general
flight path becomes too involved to be of practical use. The solu
tion to the hypothetical linear climb path, however, not only is
relatively simple but also lends itself to ready application in field
operations. The final result of the analysis of this problem is
contained in the two equations,
V 1+
m
where m is the slope of the climb path and D is the length of the
field. The first equation determines the stake position a; the second
permits the maximum altitude for that particular stake position
to be computed. The slope of the climb path may be obtained in a
trial flight from the angle of climb 6 which may be measured from
the ground with a surveyor's transit, a sextant or an inclinometer.
The slope may be measured directly with a highway engineer's
inclinometer graduated in per cent of grade. To relieve the user of
this technique of the drudgery of computation the following table
has been computed for various angles of climb.
SHOWING STAKE POSITION AND ALTITUDE FOR VARIOUS
ANGLES OF CLIMB
?n
Angle
o
10
o
.18
n
D
o
.19
20
.36
.31
25
3o
35
40
45"
47
.58
-70
.84
i-o
.38
47
45
49
.50
a
D
.50
.41
.33
.29
.25
-22
.19
0.15
LAUNCHING METHODS
95
This technique for gaining a few hundred more feet of altitude

from any small field is simplified further by the use of a direct read
ing meter. The meter is easily constructed by drawing the diagram,
"Auto-pulley Tow Meter," to scale on a piece of cardboard. A small
weight, acting as a plumb bob, is hung by a thread from a pen at
AUTO-PULLEY Tow METER

the center O of the meter. In use the observer stands at a point over
which the glider begins its steepest climb and sights on the glider
along the top side of the meter. At the peak of the glider's climb
h
a
a second observer reads the value of and as indicated by the
weighted thread on the two scales of the meter. The numerical
values so obtained when multiplied by the field length D determine
the stake position and the corresponding maximum altitude. In
this technique no corrections are necessary for wind velocity since it
enters into the climb path slope measurement. The various dis
tances may be measured to sufficient accuracy with the tow car's
odometer.
WINCH TOW
During the past few years winch towing as a means of launching
gliders has rapidly grown in favor until it is now recognized as the
most efficient method for pilots past the '"B" license or 360 turn
stage. With this system an engine-driven drum winds up the rope
and replaces the towing automobile to accelerate the glider to flying
speed. Two of its advantages are very quick acceleration and per
fectly smooth towing. Another is the fact that only the glider and
96
the drum are accelerated, which puts considerably less load on the
engine as compared to auto tow where the whole car has to be
brought up to the flying speed of the glider. Sometimes, also, higher
tows can be made than are usually possible by auto tow because
the towrope or wire can be laid across rough ground unsuitable for
the tow car.
Given a smooth field as a launching area the maximum height
possible is approximately the same for winch as for auto tow. Using
3500 feet of towline on a runway the same length with a 5-10 m.p.h.
wind the average glider can be climbed to about 800 feet before
having to be released. The same height can be reached by auto
tow with a looo-foot length of towline, the car traveling 2500 feet.
A variety of winches has been built by different glider clubs.
Most have been successful and many have incorporated new and
useful devices. However, no standard type has yet been built with
all the good features of these winches so the best of them will be
described. Probably the first successful glider launching winch in
America was that built by Gustave Scheurer at Millington, New
Jersey. It was also the simplest design and the forerunner of many
built since.
Using a Model T Ford as motive power, a drum was bolted to
the right rear wheel. That wheel was jacked up and the other
wheels were blocked. Towing was done in high gear and the towline
was wound on smoothly by being guided with two sticks held by
someone standing beside the car. Shortly afterwards the Y Flying
Club of Newark, New Jersey, built a similar winch on a 1919
Dodge sedan using a 16-inch diameter drum with flanges made of
14 -inch metal plates bringing the outside diameter to 30 inches.
Wooden pulleys, mounted on bicycle front wheel hubs, were used
for guide rollers. The side rollers were steel tubes mounted on Ford
generator bearings. This design was later improved by changing
the drum to the left rear wheel with the rope feeding through a
second set of guide rollers on the left front fender. The first set of
rollers was mounted on a level winding, hand-operated device placed
just behind the driver's seat so that he could operate it when a
second person was not present. The principal objection to a winch
with drum mounted on the rear wheel of a car is that the unusual
load exerted on the wheel bearing on that side due to the differ
ential of the rear axle is apt soon to wear out the bearings.
The main essentials of a glider winch are an engine-driven drum
which will hold 4000 to 6000 feet of % G -inch manila rope, a level
winding device, guide rollers, a rope cut-off device, and a brake to
LAUNCHING METHODS
97
stop the drum. This should be of such diameter that towline speeds
up to 45 m.p.h. are possible without excessive engine speeds.
The ideal level winding device is one automatically operated by
a worm gear connected to the driving mechanism. With such a
device properly geared one can be sure of smooth winding of the
towline without having to pay any attention to it. To save cost,
however, most winches have been built with a manually-operated
winding device. One of the best examples of this is the system used
on the Meeker winch built in Detroit. With this compact unit,
evolved from a Model A Ford, and moved about as a trailer, the
operator turns the erstwhile steering wheel of the car to turn the
winding device as he sits facing away Irom the engine and toward
THE SCHFURER WINCH ON MODEL T FORD IN 1928

the glider being towed. Although it gives the winch operator more
to do and think about, this type of manual winder is preferable to
the hand type which is more generally used and requires another
man to operate it.
Another interesting solution of a one-man winch using a manual
winder is that built by the Purdue Glider Club. In this winch the
drum is mounted directly on the drive shaft of a Model A chassis.
The drive shaft has been discontinued from turning the rear wheels,
so this winch must also be moved by towing as a trailer. A throttle
has been mounted within easy reach of the operator who stands
beside the drum where he can shift gears, use the brake and push
back and forth the level winder equipped with guide rollers.
If possible the guide rollers should be at least 4 inches in diam
eter to avoid excessive rotational speeds. They should be made of
steel rather than bronze so that a wire towline also can be used.
98
Wire will soon cut grooves into bronze rollers. The chief advantage
of using wire on a winch instead of rope is when operations arc
being carried out on beaches where there is loose, wet sand. Wet
rope soon picks up enough sand to more than double its weight,
making high tows impracticable. The sand also has a very destruc
tive effect on the winch bearings which are exposed to it. Another
advantage, which is sometimes more than outweighed by the
troubles of kinking and frequent breakage is that higher tows are
possible due to the minimum aerodynamic drag. Tests with a sail
plane of 700 pounds gross weight on a 4ooo-foot field in a 15-20mile wind and a climbing airspeed of 42 m.p.h. resulted in a
maximum altitude before release of 1500 feet using a wire towline.
"a;'tf^'"'""*v' "
Fred T. Loomi.s
THE DU PONT WINCH LAUNCHES THE "ALBATROSS" AT ELMIRA

This was approximately 300 feet higher than possible with 3/g-inch
manila rope.
The rope cut-off device, usually called the guillotine, is essential
for safe towing operations which must always allow for a mean of
detaching the towline at the towing end in case of a failure of the
glider release. Mechanical failures of approved type of releases have
been very rare, but there have also been human failures where the
pilots forgot to release and were saved from serious and perhaps
fatal accidents because the winch operator was able to cut the rope.
Lacking such a device some gliding clubs have a man standing
near the drum with a hatchet. For wire towing large pliers have
replaced the hatchet. Quick twists in opposite directions will easily
snap wire. However, these are no better than makeshifts and can
LAUNCHING METHODS
99
never replace a well-designed guillotine for quick, sure severing of

the towline.
One type of guillotine uses a single knife blade held up hori
zontally against the pull of two strong springs by a simple trigger
device. When the operator pulls the string attached to the trigger,
the blade is pulled down by the springs against a solid metal block
against which the towline is cut. Another excellent guillotine was
that developed by E. Paul du Pont on his winch. This consisted
of twro blades mounted behind the guide rollers. Strong rubber
bands acted as springs, the blades being mounted in such a position
that their leverage was powerful enough, when tripped, to snap a
broom handle placed between them. This winch was later changed
and the guillotine made like the French executioner's machine in
THE M. I. T. WINCH
having the knife made very heavy and its weight, when released,
being accelerated downward by heavy rubber bands giving the
same cutting power as the first type. With a cutting device of such
power there is real danger of serious injury for the careless and
inquisitive onlookers; prominent signs should be placed to warn
them away when the winch is ready to tow. When not in operation
the knife should be left down or blocked so that it cannot be
tripped accidentally.
A brake for the drum of a winch is absolutely necessary. In most
winches acquiring their power from the rear wheels of a car this
essential is supplied by the foot brake of the car. In winches like
the Brown-Woodruff where a separate power unit drives the winch
a shoe brake is mounted on a small drum on the axle of the winch
drum.
One of the most successful winches developed in recent years is
100 FLIGHT WITHOUT POWER

the one built by the Aeronautical Engineering Society of Massa
chusetts Institute of Technology in Cambridge under the direction
of Parker Leonard and Karl Lange. A La Salle sedan was stripped
to the chassis as far forward as the driver's seat. The drum was
mounted over the rear wheels and actuated by friction from both
rear wheels through a second set of wheels and tires mounted at
either end of the drum shaft. When operating built-in jacks raise
the rear wheels clear of the ground. When the winch is being driven
cross country the drum and its wheels are jacked up away from
them. An automatic level winding device is built above the drum
and the towline runs forward through a second set of guide rollers
mounted above the windshield. The operator faces in the direction
of the glider and can use the throttle, clutch, gear shift and brake
of the car to control the winch. A copy of this winch was built on
an Auburn sedan chassis by the Airhoppers Gliding and Soaring
Club on Long Island.
The same rules about towlines as described under AUTO TOW
hold true for winch towing, with the one exception, of course, that
length of towline unwound from the drum is determined only by
the extent of the operating field. A weak link of about 600 pounds
strength (^-inch rope) should be used ahead of the shock link on
the glider end when using wire. The parachute is also necessary.
When using rope no knots of any kind should be permitted. End
connections can be made by a short splice of not less than 6 tucks.
Eye connections to rings should be protected with thimbles and
eye splices made with at least 3 tucks.
The operation of a winch requires more skill than a tow car driv
ing and should be done only by or under the close supervision of
some one thoroughly familiar with it. The tow must be started in
the gear to be used throughout the tow as the drum does not have
enough inertia to permit gear shifting without danger of fouling
the line. The choice of gear to be used depends on the power and
pick-up of the engine, the gear ratio of the engine to the drum, the
weight and the drag of the glider and the wind velocity.
A set of flag signals should be used in winch towing and these
should be thoroughly understood by everyone taking part in the
launching. A good scheme is to use red and white flags about 3 feet
square on light sticks or poles 6 feet long. Some clubs use a red flag
mounted on the winch or stuck in the ground near the glider,
denoting that things are not yet ready at that end of the line. When
the winch operator is ready to tow he substitutes the white flag for
the red. When the glider pilot has fastened his safety belt, closed
his cockpit cover and is ready to be launched, he calls to the man
LAUNCHING METHODS
101
holding his wing tip. This helper then holds the wing tip level and
cither he with his other hand, or someone else if available, holds
the white flag over his head and waves it slowly from side to side.
This is the signal to take up slack which may have to be repeated
by a second signal man halfway to the winch if the tow is very
long or over a slight elevation.
The winch driver, having previously given his ready signal denot
ing that the winch engine is running and has been warmed up, lets
out his clutch slowly to take up the slack gradually. When the ship
begins to move the signal man at the glider drops his flag and the
winch operator gives the engine full throttle to accelerate the drum
as quickly as possible to the proper towing speed for the wind
cjimt> a4>os^ /OO feet
eJ~ f-irsf"
WINCH LAUNCHING
velocity at the time. It is helpful to have a wind sock or wind
velocity device of some kind such as an anemometer mounted on
the winch so that the operator may keep posted on the wind direc
tion and velocity at all times.
A pilot experienced in winch towing takes off in a gradual climb
until he is about 100 feet high. This is to allow the winch operator
to get the drum up to the speed where the engine will have enough
power to handle the climb as well as to prevent the danger of a
stall resulting from towline breakage too low to recover. Above this
height he may pull back quite steeply to get the maximum height
possible from the tow which is much smoother than auto tow, per
mitting a steeper climb without undue strain on the glider. As the
glider changes to the steeper climb the winch should be slowed
down to keep the glider air speed at the correct velocity. Too fast
a tow will prevent a maximum climb and will overload the wings
of the glider. The final speed near the end of the tow usually will
be about one half the maximum unless the wind velocity tends to
increase with altitude, in which case it may be less.

The pilot should watch his airspeed indicator closely and may
use an arm signal to indicate to the winch operator whether he is
being towed too slowly or too fast. Sudden variations in the angle
of climb should be avoided as it is difficult to vary the towline
velocity to allow for these changes. Drifting off to the side should
also be avoided. If the wind has changed direction slightly so that
the tow is crosswind the glider can be kept in a straight line toward
the winch by a slight crabbing accomplished by holding a certain
amount of rudder in the direction of the wind.
As in auto tow, the pilot should level out before releasing the
towline. If he waits too long before releasing he may start to be
accelerated downward which will give him a false feeling of exces
sive flying speed when in fact the glider may actually be stalled.
A turn made under this condition is likely to result in a spin.
The winch operator should avoid this by gradually slowing up the
drum as the towline reaches this angle. If the glider pilot hangs
on a bit too long the winch operator should bring the drum to a
full stop. If he still does not release after this the operator should
trip the guillotine and cut the towline.
The moment the pilot releases the towline the winch operator
must throw out the clutch and pull on the brake. Retrieving the
towline for the next launching is usually done by towing it back
with a car while the winch clutch is out. The car should be driven
back at a very even speed of not faster than 10-15 m.p.h. as other
wise the towline may backlash and be damaged, causing failures
under strain. The winch operator must be ready to apply the brake
on the drum as the car stops. It is usually well to drive about 10
feet past the glider before dropping the end of the towline if it is
rope, due to its elasticity which may make it pull back. The
retrieving car should be slowed down gradually and brought to a
full stop before dropping the rope.
Rope towline should never be left on the drum after the last tow
but should be unwound completely from the drum after the tow
and rewound without tension. It should be protected from rain
and dew which will tend to shrink it. It should be inspected fre
quently for weak spots, and these cut and spliced.
SHOCK-CORD LAUNCHING
Launching by means of a rubber-rope or shock-cord catapult is
not nearly so frequently used now as formerly. However, it still
remains as the only means of launching for slope soaring from the
top of ridges where the take-off areas are too limited for auto or
winch tow methods. There are two types of shock-cord launchings.
LAUNCHING METHODS
103
In the first the elastic rope is stretched by man power, and in the
second by an automobile.
The shock cord is a .%-inch diameter bundle of elastic rubber
bands which have been stretched and covered with a loosely woven
binding keeping the strands stretched about 100% of their natural
length. This cord requires about 375 pounds' pull per strand to
double its length when new. For gliders of less than 400 pounds
gross weight two strands of about 100 feet long are used. For
heavier ships a double cord (four strands) is used.
For a safety measure, about 35 feet of i^-inch manila rope should
be served into the vertex of a shock-cord V so as to isolate the glider
and pilot in case of cord failure. A welded steel ring of at least
2-inch diameter is fastened to the end of this rope by means of
=>>
HAND SHOCK-CORD LAUNCHING
thimble and eye splice thoroughly taped. When ready to launch
this ring is slipped over the open hook located on the nose of the
glider below and behind the closed hook release mechanism used
for the other types of launching. Its action is automatic, the ring
dropping free as the glider overtakes the shock cord and the slack
comes in the cord at the end of the pull.
The cord eye should be served with a galvanized iron thimble to
prevent a sharp bend in the cord. The whole joint should be pro
tected from abrasive wear by a covering of friction tape. The two
loose ends should have about 30 feet of i/-inch rope served onto
each if the cord is to be used for auto-tow shock-cord launching.
Hand shock-cord launching requires a set-up which varies with
the contour of the take-off area. When this is level and flat over the
whole area, the cord is laid out in V form so that the glider wing
tips will just clear the cord crew. In this type of launching it is
necessary to store up enough energy in the cord by stretching it by
the crew so that upon release of the tail anchorage of the glider
it will be catapulted to a velocity somewhat above flying speed

before the end of the take-off area is reached by the cord crew. It
is important to be careful not to set the glider so far back from
the edge of the hill that it will get dangerously low before it reaches
the area of lift. The set-up must be kept as far out toward the
edge of the slope as possible and still have room for an effective
launching. In the case where the take-off area is sloping ground and
the glider is sure to go above the crew, the two lines may be brought
in so that they are within 10 or 15 feet of each other. Four to seven
men may be used on each strand of cord.
The tail of the glider should be provided with a ring or hook
for the hold-back rope. One of the best ways to anchor the glider
Elmira Star Gazette
SHOCK-CORD LAUNCHING OF FRANKLIN UTILITY GLIDER

while the shock cord is being stretched is to have one end of the
rope tied to a car or trailer of sufficient weight, the rope slipped
through the ring or hook on the tail and have one or two men
sitting on the ground holding the other end.
The glider pilot gives the commands for the shock-cord launch
ing. After he is seated in the glider with safety belt fastened and
ready to take off he first makes sure that a man is holding his wing
tip level, the tail is anchored by the hold-back crew and the shockcord crew are at their posts holding the cord. He then gives the
order "WALK" and the cord crew walks forward about 12 paces,
after which he gives the order "RUN." When the crew has run about
the same number of paces, having pulled the cord to about 80% to
90% of its possible stretch, the pilot calls "LET GO." The tail crew
simply drops the rope and the glider shoots forward. The cord crew
continues to run until the glider passes over them and the cord
drops free.
LAUNCHING METHODS
105
The acceleration is so fast that the pilot has no time to think

before he is well out over the side of the mountain. He should
therefore be sure to hold his stick at neutral or even a little farther
forward to keep the glider from zooming upward. He should try
to fly straight ahead with no more than a slight climb of a few
feet. The tremendous energy of the catapult launching is short lived
and he may waste it if he pulls up too sharply. Also he may have
to fly ahead some distance into a strong head wind before reaching
the area of lift beyond the edge of the ridge.
As this very sudden acceleration in shock-cord launching is apt
to be more than a little disconcerting for a new pilot who has never
done it, no student should be launched from a hilltop until he has
had practice on a level field. It is also wise for any pilot no matter
Rope.
J5fr-
_________-SAtfcx
Cord
Kope.
7^7=-3^7"
AUTO SHOCK-CORD LAUNCHING

how experienced to try out a shock-cord launching on a level field
before attempting a launching in a sailplane new to him and from
a site not yet tested by others who are present to advise him.
Auto shock-cord launching is very similar with the automobile
taking the place of the cord crew. Flag signals are used instead of
verbal commands. The lines connected to the ends of the cord are
brought together and fastened to the release on the tow car. A
point on the ground is marked for about 50% stretch of the cord
so that the driver can signal for the tail release when the car has
passed that point. This signal is passed on by the wing tip man to
the tail crew. Less stretch is needed than for hand launching as the
mass and speed of the car are both greater than with hand pull.
Low gear should be used throughout the tow. The driver should
decide beforehand after consultation with the pilot in which direc
tion he will turn in case there is any possible danger of the glider
not clearing the car. There is also danger for the tow car driver
in case of too abrupt a turn at the end of the launching on a downslope. The forces involved in auto shock-cord launching are much
higher than in straight auto tow, so the car release and its attach
ment should have ample strength for this work.

There is a real element of danger in shock-cord launchings in
the event of the failure of any of the equipment used. If any of the
ropes, metal rings, or the cord itself should break under the terrific
strain before the glider is released it may result in serious injury.
If the break occurs at the crew or car end the force may be sufficient
to smash the nose of the glider and injure the pilot. If a failure
happens at the glider end a cord crew member or the tow car
driver may be injured. For this reason it is essential to use the best
of materials and to keep them in good condition. The cord should
be inspected frequently for any indication of wear or failure either
of the cord itself or of its fastenings. When not in use it should be
stored in a cool, dark and dry place.
AIRPLANE TOW
Contrary to common belief, airplane towing of gliders as a means
of launching is not hazardous if carried out according to proper
rules of proven procedure. Unless the air is unusually turbulent
the stresses on the glider in towed flight are actually substantially
less than in other methods of launching.
To do airplane towing in the United States one must have a
glider that has been licensed for this type of operation. This license
usually requires that the glider be placarded for the maximum
allowable speeds in towed flight. It is essential therefore that both
the airplane and the glider be equipped with airspeed indicators.
It is also required by the Civil Air Regulations that both pilots wear
parachutes. The airplane pilot must hold a Commercial Pilot's
Certificate and the glider pilot must have a Commercial Glider
Pilot's Certificate as well as a Certificate of Non-Application to the
C.A.R. ruling that no aircraft be towed behind another aircraft
except by special permission of the Civil Aeronautics Adminis
tration.
Primary gliders can never be airplane towed as they lack sufficient
strength, stability and protection. Towing is usually done in well
designed secondary or utility gliders and intermediate and high
performance sailplanes.
The ideal towing airplane is an open cockpit biplane with light
wing loading of less than 8 pounds per square foot and conse
quently low stalling speed, and between 90 and 220 h.p. Light air
planes of 40 h.p. have been used from large fields but they lack
sufficient reserve power and the required full rearward visibility.
As the best average towing speed is 50-55 m.p.h., the airplane should
have a stalling speed of not over 45 m.p.h. and preferably less.
Excellent American tow planes are the Waco F and the Fleet
LAUNCHING METHODS
107
biplanes with Warner or Kinner engines of from 100 to 145 h.p.

The airplane should be equipped with an approved release de
vice similar to that used in the glider. Complicated and costly tow
ing attachments transmitting the towing forces over the tail surfaces
and directly to the center of gravity of the airplane have been used
abroad but the American system of attaching the release directly
to the tail skid or tail wheel of the airplane has proven more prac
tical in every way. The slipstream tends to lay out the first part of
the towline directly behind the airplane regardless of the angle the
glider may be pulling on it so no trouble need be expected from the
towline fouling the rudder or elevators. The rope or cable tripping
the release can be run inside the fuselage to the pilots cockpit in a
permanent installation or around the outside for the usual tem
porary arrangement.
The towline should be i^-inch manila rope at least 300 feet long
and equipped with metal rings at each end just like the towline for
auto towing. For cross-country towing or student instruction it is
advisable to use a 4oo-foot towline if the size of the field will permit
it. On the average field at sea level and in quiet air the take-off
run of the airplane will be increased from 200 to 300 feet. Using a
cable instead of a rope towline will reduce the drag on the air
plane, but it must be supplemented with a so-foot length of 14-inch
rope, preferably at the airplane end, to give some elasticity and to
provide a weak link.
Before starting a tow the pilots of the airplane and the glider
should talk over all details of the tow and thoroughly understand
iheir signals if any are to be used. Under no circumstances should
an airplane tow be made with both pilots inexperienced in this type
of towing. A tow may be made with a capable airplane pilot who
has never towed a glider before if the glider pilot is experienced in
airplane towing. Also a safe tow may be made with the airplane
pilot experienced in airplane towing and the glider pilot new to
this type of launching if he has had complete verbal instructions.
After the glider is thoroughly inspected it should be set at the
end of the field to permit the longest tow as nearly into the wind
as possible. The towline should then be laid out straight and the
other ring attached to the airplane. After checking both releases
by tripping them with the line under tension, the airplane should
be warmed up. When ready it should move slowly ahead to take
up the slack in the line. If the pilot should run up his engine at
the last moment he should throttle back and wait a moment until
the air has quieted down before starting a tow.
When ready the glider pilot calls to the man holding his wing

tip level, who gives a hand or flag signal to the airplane pilot. In
starting, the airplane pilot should simply make a normal take-off.
The only difference he may notice is that his tail may come up more
slowly than usual to flying position and that his ground run is
somewhat longer. Once in the air he should watch his airspeed
indicator closely, keeping a constant speed not exceeding that
allowed by the placard on the glider. He should make gradual
turns. If the air is turbulent he must fly as slowly as possible, con
sistent with safety. If he is the gliding instructor of the glider pilot
he may use hand signals or rock his wings gently from side to side
indicating to the student when to release. As soon as he sees that
the glider is off he throttles back and dives down to get the rope
away from the glider. The best way for him to drop it on the
field is to dive down at 70-80 m.p.h. with engine partly throttled
so that the line will string out and up away from obstructions and
then pull the release when he is about 200-300 feet in the air over
the center of the field. All towing of beginners should be made in
quiet air.
As he begins to move forward on the take-off the glider pilot
holds his stick at neutral. As he reaches flying speed he should pull
back gently and climb to about 15-20 feet. Then he should dive
down to about half that height to allow the rope to slacken a bit
and the airplane to take off. During the climb and throughout the
tow he should fly directly behind the airplane and perhaps also
5-10 feet above its line of flight. If he drops well below he is in dan
ger of getting into the slipstream which, with all American engines
rotating clockwise, is spiraling off down and to the left. If this
happens his left wing may drop and he may have trouble bringing
it up again. If he cannot soon lift it back level he must release.
If the glider pilot should pull up too high on the take-off before
the airplane leaves the ground he may lift the tail of the airplane
and cause the propeller to hit the ground and be damaged. Pulling
up too high when the airplane is flying at low altitude is even more
dangerous as the plane can be put into a dive from which the pilot
cannot recover in time before crashing into the ground. In a well
executed airplane tow the glider pilot flies formation with the
airplane closely watching the airplane's every movement and trying
at all times to stay in the same position with respect to it. This
may seem somewhat difficult at first but becomes easy with practice.
In making a turn in towed flight the glider should be steered
toward the outside of the circle to keep the towline taut. If the air
plane is making a right turn the glider pilot holds enough left
rudder to accomplish this. The airplane must maintain power at
LAUNCHING METHODS
109
all times while towing. If it is necessary to lose altitude in tow this

must also be done with power on and gradually to prevent excessive
speed which will cause the glider to overtake the airplane.
Special care must be taken when towing on days of good con
vection when there are strong up and clown currents. If the airplane
encounters an upcurrent it may be lifted well above the glider
which then runs into danger of getting into the slipstream. When
the glider pilot sees the airplane rising suddenly above him he
should immediately pull back and attempt to stay behind it. This is
usually not hard as at towing speed it has plenty of reserve speed
for this.
If the glider is in an upcurrent when the airplane is not it will
be quickly lifted high above or will catch up with the airplane if
\ CORRECT TURN IN AIRPLANE TOWING

the pilot is not careful. A valuable way to lose height to stay behind
the airplane is to open the spoilers increasing drag and reducing
lift. The glider can also be skidded from side to side to slow it up
and take the slack out of the rope.
Both pilots should be prepared to release immediately in case
of emergency. If the airplane pilot should experience engine failure
on the take-off he must release the line at once. The glider pilot
must be ready to do the same right afterwards, as attempting to
land with the line dragging may cause it to snag in trees or other
obstructions. The glider pilot may suddenly find himself in a vio
lent upcurrent of such turbulence that his craft is becoming dan
gerously stressed and in this case he should release. Usually he is
given warning of approaching conditions as he sees the airplane
encounter them first. No towing should be done in a glider that

is incorrectly rigged so that there is wing heaviness on one side.
If a pilot should find himself being towed in a glider in this con
dition he should cut loose as soon as he is high enough so that the
towline trailing behind the airplane will not foul in anything on
the ground, and to enable himself to make a safe landing.
CHAPTER VI
SOARING METEOROLOGY
By Dr. Karl O. Lange
I HE POWER REQUIRED for soaring is not derived from

any power plant carried by the craft, but has to be taken directly
from the energies available in the atmosphere. At our present state
of experience, the atmospheric energy is used extensively only in
the form of upcurrents. These upcurrents are wind currents which
have an upward component.
Soaring meteorology is chiefly concerned with the study of up
currents of a size that can be utilized in flight. It points out the
possibilities and dangers of the atmosphere as a source of energy.
A fundamental knowledge of these meteorological facts is required
of any glider pilot who hopes to be successful.
Glider pilots differentiate among several frequently encountered
kinds of upcurrents in the atmosphere:
1.
2.
3.
4.
5.
6.
Dry thermals
Slope winds
Cumulus clouds, including heat thunderstorms
Cloud streets and waves
Squalls and fronts, including front thunderstorms
The "Moazagotl" condition
To understand them, it is necessary first to get acquainted with

some general characteristics of the lower atmosphere.
The Temperature of the Atmosphere
The temperature of the earth and the atmosphere are the result
of radiation received from the sun and radiation sent out from the
earth into space. The amount of sun's radiation received at a par
ticular location depends on the position of the sun; in other words,
on the latitude, the season, and the time of day.
The sun's rays enter the earth's region at the outer atmosphere.
If the atmosphere is clear, they penetrate the transparent medium
without losing more than 20 or 30% of their energy until they hit
the earth's surface. Here, they are partly reflected but mostly ab
sorbed and converted into heat, thus making the temperature rise,
111

If the sky, however, contains clouds or haze, then a certain per
centage of the sun's radiation is reflected, dispersed, and absorbed
during its passage through the atmosphere, and only the remaining
part can heat the earth's surface. Thus the diurnal rise of tempera
ture on the ground is less pronounced on such days.
The earth's radiation into space takes place all the time. It tends
to lower the surface temperature, and is particularly noticeable by
this effect at night when not compensated by solar radiation.
The intensity of radiation depends on many factors, notably on
the temperature. The energy sent out from any body is proportional
to the fourth power of the absolute temperature T
E = const. X T4
The absolute temperature is 273 + t, where t is the surface tem
perature in Centigrade. For example, if the ground temperature
of a certain location is 100 F., more than twice as much heat is
radiated off as when the ground is only 10 F.
On clear nights this radiation leaves the earth and little returns
from the atmosphere, causing the temperature to drop. "Clear cold"
nights are proverbial. However, if the sky is covered with a cloud
layer or if there is fog or haze, large amounts of the earth's radiation
are reflected back and absorbed. Consequently the night does not
become so cold as it would have had it been clear. We see that
clouds or fog or even haze again tend to diminish the diurnal
change of temperature on the ground.
These considerations show that the character of the sky plays a
role in determining the temperature on the earth. They show too
that the temperatures of the atmosphere are to some degree directly
affected by radiation. The warming up process during the day starts
and is most pronounced at the ground. The same is the case with
cooling at night. More often than not, a diurnal temperature varia
tion of 20 at the ground is diminished to nothing at the 6000 foot
level.
Generally speaking, we find the highest temperatures at the
earth's surface. From there, heat is transferred into the higher layers
of the atmosphere partly by conduction, but mostly by a turbulent
exchange of air, and by condensation processes. The result is that
ordinarily the temperature decreases rapidly with altitude. The rate
at which the temperature drops depends on many factors, notably
the time of day and the season and the location on the earth's
surface at which the air mass is found or over which it had orig
inally formed and traveled. The actual temperature conditions in
the atmosphere show great variations, but as a first approximation
SOARING METEOROLOGY 113

and for many purposes the temperature can be considered to fall
off on the average at a rate of 3 F. per 1000 feet.
The rate at which the temperature changes with altitude is called
the temperature lapse rate. The atmosphere is said to be stable or
to have a stable lapse rate if the rate of change is less than 5.5 F.
per 1000 feet. If it is 6 F. or more per 1000 feet, the atmosphere
is called unstable. If the temperature does not change vertically,
the condition is called isothermal, and if the temperature increases
with altitude, we have an inversion. If such a layer in which the
temperature increases with altitude starts right at the earth's sur
face, it is called a ground inversion. The lapse rate is usually rep
resented by a curve in a diagram where temperatures are the
abscissae and where altitudes (or logarithms of air pressure) are the
ordinates. Fig. i depicts the various possible lapse rates in the
atmosphere.
1000
Meters
1500
\
1000
500
5C
\\
\\
\\
15
FIG. i. TEMPERATURE LAPSE RATES IN THE ATMOSPHERE

A = ground inversion, B = stable lapse rate, C = isothermal condition, D =
superadiabatic lapse rate, E = adiabatic lapse rate, F = inversion, G = stable
lapse rate. The dotted lines represent adiabatic changes of state as they occur
in "dry thermals."
The vertical temperature distribution in the atmosphere is not

only important for soaring, but is connected with many other
weather phenomena. For this reason, the daily determination of
these conditions all over the country now forms an important part
of the routine weather observations. The U. S. Government deter
mines the atmospheric temperatures early every morning at more
than thirty points scattered all over the United States by means of
radio-meteorgraph (radiosonde) ascensions. These measurements are
114 FLIGHT WITHOUT POWER.

made and used in the metric system. The temperature is expressed
in Centigrade, the altitude in meters, and the pressure in millibars.
Five Centigrade degrees is equal to 9 Fahrenheit degrees, and o
C. corresponds to the freezing point, 32 F. One hundred meters is
approximately 330 feet; 1000 millibars is equivalent to 29.5 inches
of mercury. These radiosonde ascensions are now undoubtedly the
most valuable material for the determination and forecasting of
soaring conditions. When used with due consideration of the
diurnal changes which take place in the lower layers, the soundings
give a true picture of the energies available for soaring, as is shown
in the following pages.
The Air Pressure
The air pressure is the result of the weight of the atmospheric
gases. (On a winter day, a cubic meter of air at the surface weighs
almost 3 pounds.) Naturally, the air pressure is always highest at
the ground, because the entire atmosphere lies on top. At a higher
elevation, part of the air is left underneath, and only the air on
top exerts its weight. Consequently the air pressure decreases with
altitude.
The rate at which the air pressure decreases with height depends
on the density of the air. The density in turn depends on air
pressure, air temperature, and moisture content. Since these fac
tors vary from day to day and from location to location and from
height to height, the exact rate of pressure change with height must
vary too. In order to obtain accurate correlation of pressure and
height, it is necessary first to measure all temperatures, humidities,
and pressures from the ground up. Thus all air densities from the
ground up can be determined. By integration, it is then possible
to compute the pressures which are found at certain altitudes.
These computations are carried out every day at the radiosonde
stations and the results may be obtained from the U. S. Weather
Bureau, if needed for the exact determination of altitudes from
barographs or for check-up on altimeters. If a smaller degree of
accuracy is sufficient, the data may be taken from tables which are
computed for the "standard atmosphere." The standard atmosphere
assumes a constant temperature and a constant pressure on the
ground, and a constant, stable lapse rate of 3.57 F. per 1000 ft.
Air Density and Lift
The density of ideal gases is given by temperature and pressure.
The air consists of a mixture of nitrogen, oxygen, carbon dioxide,
water vapor, and small quantities of other gases. All except the
SOARING METEOROLOGY
115
water vapor behave as ideal gases under ordinary atmospheric con

ditions. Therefore, when neglecting the presence of water vapor in
the atmosphere for the present, we can say that the air follows the
gas laws. The density of dry air is
^J I . 1 O r-|V
where
Q = density in kg per cubic meter

p = pressure in inches of mercury
T = absolute temperature.
For example, at a pressure of 30 inches and a temperature of
32 F., the air density is 1.293 k Per cubic meter. At the same
pressure, but at a temperature of 86 F., it would be only 1.178 kg,
more than 10% lighter.
If a body is submerged in a liquid, it receives a lift which is equal
to the weight of the liquid which it replaces. The weight of the
body acts contrary to this lift. Consequently, if the density of the
body is higher than that of the liquid, it sinks. If the body is
lighter than the liquid, it receives a free lift. The lifting force is
proportional to the density difference.
This law holds for gases as it does for liquids, as demonstrated
by airships and balloons. That means that a quantity of air con
tained in air of a higher density receives a free lift, and it holds
true whether the quantity of air is held together in a hot air balloon
or is just an air bubble, the temperature of which is higher than
the surroundings.
So long as the air is not contained in a closed balloon, like a
rubber toy balloon, the air pressure inside and outside will always
be equal. Thus the density difference, which determines the free
lift, can simply be expressed in terms of temperature. The upward
acceleration created on a quantity of air of a temperature T, sub
merged in the atmosphere at a temperature T0 is
T
-*- T
* o
= g^
AO
g is the gravity constant

T is expressed in absolute degrees
For example, if on a summer day with a temperature of 300
(81 F.), a quantity of air blows slowly over a particularly hot
spot on the ground, getting heated up to 301 (83F.), it would
experience an upward acceleration of 3.2 cm./sec./sec. This is not
much, only about one third of a per cent of the gravity accelera
tion. However, if the temperature difference were maintained long
enough, as the "thermal" moves upward, the upcurrent would soon
be accelerated to appreciable vertical velocities. If the thermal rises.

under a steady acceleration b its vertical speed Vz at a height H
would be Vz = \/2bH
Friction between the upcurrent and the undisturbed atmosphere
reduces the speed by 20% to 30%, so that
V* = 0.75 X
For example, at a height of 100 feet, the upcurrent would be about
i meter per second. At 1000 feet it would have grown to about
3.3 meters per second.
These considerations show that there are two requirements for
strong thermal currents. Either the temperature difference must be
very large, which is seldom experienced, or the temperature differ
ence between thermal and surroundings must be maintained over
a considerable range of altitude to permit building up of high
vertical speeds.
The Temperature Variation of an Ascending Current
If a gas expands from one pressure to a lower one, its molecules
have to be distributed over a wider space. The energy required to
do this is taken out of the gas in the form of heat, thus lowering
the temperature of the gas. Similarly, when a gas is compressed, its
temperature rises.
A quantity of air which moves upward in the atmosphere comes
under lower pressure. That is, it expands and its temperature
drops. If the current moves downward, the opposite takes place.
The air temperature rises. Provided that no heat is added or sub
tracted during the process of vertical movement, then the tem
perature of the vertical current changes almost exactly 1 C. per
100 meters. Such a process, in which no exchange of heat with the
surroundings or by radiation takes place, is called an adiabatic
process. And a lapse rate of 1 C. per 100 meters is therefore called
an adiabatic lapse rate.
All slope currents and all "dry thermals" follow this law. There
is naturally a certain amount of turbulent mixing along the edges
of the upcurrents, which tends to establish a gradual change from
the temperatures of the current to those of the surroundings. The
inside of the vertical current, however, changes 1 C. per 100 meters
(approximately 5.5 F. per 1000 feet) no matter what the initial
temperature is, nor at what altitude the process takes place.
The Lapse Rates
Comparison of this fixed adiabatic lapse rate of the vertical cur
rents with the lapse rates present in the atmosphere, as measured
by radiosonde ascents, forms a criterion of the conditions for soar-
SOARING METEOROLOGY
117
ing. Fig. i shows various possible lapse rates. The dotted lines
show the adiabatic lapse rates.
From the ground up, there is indicated first a ground inversion
marked A, starting with a surface temperature of 20 C. Suppose
an upcurrent got started somehow, perhaps by the air being forced
over a ridge. The upcurrent must vary its temperature according
to the dotted line starting from 20. Immediately the temperature
of the upcurrent becomes considerably lower than the ambient
temperature. At an elevation of 200 meters, the temperature is
already about 5 C. That means the air that was forced up is con
siderably heavier than the surrounding air. It has a tendency to
fall back to the surface, or rather in reality it would never get that
high as it would rather flow around the ridge than over it. Ground
inversions are encountered most frequently on calm clear nights.
There can be no soaring in the altitude range covered by the
inversion. All vertical air movements are choked.
Condition B in Fig. i indicates an extended stable lapse rate. If
an upcurrent would get started at the highest temperature, it would
follow the second dotted line. It is easy to see that again a negative
temperature difference results. All upcurrents which might get
started in the range of altitude from 200 meters to 800 meters on
the graph would soon assume a lower temperature than the sur
roundings. All downcurrents would become warmer than the sur
roundings. While the choking is not so severe as in the inversion,
it is sufficient to suppress or greatly diminish vertical movements.
Stable lapse rates are common in "bad weather"; that is, in steady
rain or when layers of stratus clouds are present. Soaring is very
limited.
The isothermal layer C acts much like an inversion. Upcurrents
that reach into it are quickly suppressed. Perhaps the best known
isothermal region of the atmosphere is the stratosphere. Vertical
exchange of air and moisture cannot penetrate deeply into it.
Hence the lack of bumpiness and clouds in the stratosphere.
Condition D on Fig. i indicates a superadiabatic lapse rate, a
theoretically ideal soaring condition. If the slightest vertical motion
got under way anywhere in this region, the result would be an
immediate increase in temperature difference, which is now posi
tive. The higher the upcurrent goes, the lighter it becomes with
regard to the surroundings. Its acceleration rapidly increases as it
rises. Unfortunately the superadiabatic lapse rate is an unstable
condition which cannot extend over a considerable range of height
nor last very long. It is found near the ground, when intense heating

of the ground takes place. At greater heights it is a rather rare
phenomenon.
The lapse rate marked E in Fig. i is adiabatic. If a quantity of
air in this layer somehow gets a push upward, it continues upward
until the original impetus is used up by friction, because its tem
perature is the same as the outside temperature at all levels. The
upcurrent is in equilibrium with the surroundings at all heights
and no thermal forces tend to restrict the vertical motion at all.
A "thermal" (a quantity of air which somehow got a little warmer
than the surrounding air) stays warmer at all heights. The upcur
rent is continuously accelerated upward until the excess tempera
ture is dispersed by lateral mixing or until the adiabatic layer is
traversed. Adiabatic lapse rates occur frequently in the atmos
phere, near the ground as well as at greater heights. They are the
rule on clear and windy days.
Fig. i contains an inversion F on top of the adiabatic lapse rate.
If an upcurrent has formed underneath and has traveled through
the adiabatic layer, it will push into the inversion on account of its
inertia. As it proceeds, however, an increasing temperature differ
ence between upcurrent and surroundings is formed. In other
words, a downward force is now acting on the updraft, which
makes it spread out horizontally and chokes it. Inversions limit the
extent of altitude soaring. As a rule it is not possible for upcurrents
to penetrate a well-defined inversion, and consequently one cannot
soar through inversions. There are cases, however, where upcurrents
are so warm and strong that they rupture a small and even mod
erate inversion and emerge with a temperature that is still in excess
of that of the layer of air above the inversion, as is often the case
with thunderstorm currents.
The Diurnal Variation of Soaring Conditions near the Ground
The preceding paragraph shows the dependence of vertical air
currents on the temperature lapse rates of the atmosphere, and it
has been pointed out that the temperatures, especially those on the
ground, are the result of radiation. During a twenty-four hour
period, we go through one cycle of radiation, from maximum sun
shine around noon to no sun at night to maximum sunshine around
noon of the following day. Consequently the temperature variation
on the ground goes through a twenty-four hour period with maxi
mum temperatures sometime after noon, minimum temperatures
after midnight to maximum temperatures again shortly after noon
of the following day. The temperature variation on the ground
causes a twenty-four hour variation of lapse rates in the lower
SOARING METEOROLOGY
119
layer of the atmosphere. In turn, chances for the formation of

vertical currents vary on a twenty-four hour basis.
Figs. 2 and 3 give an indication of what occurs during a one-day
period in the lowest 1000 to 3000 feet. It must be realized, how
ever, that there are many other weather factors besides radiation
which may superimpose on this picture and alter it considerably.
The considerations apply under the assumption that the same
type of weather is preserved for twenty-four hours. Two cases are
considered: a clear calm day and a clear windy day.
4AM
SAM
9AM
NOON
2PM
4PM
7PM
IIPM
FIG. 2. THE DIURNAL VARIATION OF THE LAPSE RATE IN THI

LAYER NEAR THE GROUND ON A CLEAR CALM DAY
At night there is a ground inversion; after sunrise the ground inversion be
gins to dissolve; in midday an adiabatic and, nearest the ground, a superadiabatic lapse rate forms; during the afternoon the air near the ground cools,
aloft an adiabatic lapse rate remains; in the evening a new ground inversion
develops.
The aerological radiosonde ascensions are carried out at 4:00 A.M.

In the great majority of cases they reveal that at that time of the
morning a ground inversion exists, as depicted in the first sketch
of Fig. 2. A few hours later the sun rises and begins to warm the
ground. The thin film of air directly over the ground receives heat
from the ground and tiny eddies and whirls begin to form. Thus
air particles that were in contact with the warm earth are carried
upward and mixed with the cold air there. New heated elements
follow, and by and by a layer is formed which has an adiabatic
lapse rate and through which more and more heated air is ex
changed from near the ground to the upper part of the inversion.

But the free exchange of currents from the ground to the upper
layers of the atmosphere remains handicapped until the last of
the ground inversion is wiped out. During the June-July Soaring
Contests at Elmira, New York, it usually is 9:00 A.M. to 10:00 A.M.
or even later before this is the case. Only from that time on do the
thermals and ridge currents shoot up from the valley to great
heights.
Then the energy of insolation serves to heat up the whole "turbu
lence zone." The lapse rate remains adiabatic and may even become
superadiabatic and the mean temperature of the entire lower region
of the atmosphere increases, as shown in the sketch for 2:00 P.M.
The maximum temperature of the day usually occurs at around
2 P.M. From then on the earth's surface begins to cool off as the
sun declines. By about 5:00 P.M. the effect is quite noticeable in
the atmosphere. The ground becomes cooler and so does the film of
air that is in contact with it. That means that this air now becomes
relatively heavy. It clings to the ground and resists being carried up
by the turbulence and mixed with the higher layers. The cooler it
gets, the more tenaciously it adheres to the ground. A new ground
inversion then forms. At the same time, thermal conditions above
may remain quite good. It is frequently observed that pilots con
tinue to soar at heights of a few thousand feet after sundown, while
it is impossible for others to get up to them from the ground.
During the night the ground inversion intensifies under the
influence of the earth's radiation. And during the following day
the cycle repeats itself unless a change of weather occurs. Fig. 4
shows actual air movement over an airport. A number of floating
balloons were released from an airplane and their paths triangu
lated from the ground. All the balloons which were followed in
the early morning hours went almost parallel with the ground. In
contrast, the balloon of 11:30 A.M. shot up in a thermal of 500 to
600 feet per minute.
The diurnal variation of soaring conditions is different when a
strong wind prevails. Fig. 3 illustrates what happens on such a day.
At noon and shortly afterwards, the picture is similar to that of a
calm day. There is an adiabatic layer from the ground up. When
the ground begins to get colder, again a thin layer of cold air wants
to form over it. But the strong wind with its turbulent structure
picks up the cold air and carries it into the higher layers, all the
time mixing it with the air which is already there. The result is
that the turbulence preserves an adiabatic lapse rate and the cool
ing effect is distributed over the entire "turbulence zone." The mean
temperature then drops. All this time energy has to be spent in lift-
SOARING METEOROLOGY
121
ing the heavy cold air off the ground. The energy is taken from the
wind, which results in a decrease of wind velocity. Moreover, as the
process goes on, the turbulence zone shrinks in height. Therefore,
if an attempt is to be made to soar all night, a weather situation
should be chosen in which the pressure gradient is large enough
to guarantee that the wind will not die down during the night.
Furthermore, a low ridge is preferred in order to be sure that it
stays in the turbulence zone and does not get into the inversion on
top of it. The diurnal temperature drop is less pronounced over
water. Thus, chances of preserving an adiabatic lapse rate during
the night are best for an on-shore wind.
4PM
IIPM
4AM
FIG. 3. THI VARIATION OF THE LAPSE RATE ON A CLEAR

WINDY DAY
In midday a high unstable layer exists; during the afternoon and evening, the
mean temperature of the turbulence zone drops and the turbulence inversion
becomes lower.
Thermals
It was shown that the temperature lapse rate governs the develop
ment of vertical currents once they get started. The initial upward
movement is most frequently caused right at the surface of the
earth, but at times also in greater heights. It may be caused me
chanically or thermally; that is, we get slope winds or thermals.
Frequently the two kinds form simultaneously or combine.
When the sun shines it heats the ground. Patches of good ab
sorption and low heat capacity rise to high temperatures as, for
instance, a desert or, on a smaller scale, any sandy spot. Other loca
tions assume other temperatures according to their physical prop
erties. If the ground is wet, a good portion of the energy received
from the sun is spent on evaporation. It is well known that the
diurnal temperature rise is delayed on mornings with dew. We
know that woods stay cooler during the day, and ocean and lake
regions offer coolness in spite of intense summer insolation.
Ground temperatures of 200 F. have been measured while the
air temperature, determined at 6 feet above ground, was less than

100. This shows that the temperature gradient in the lowest 6 feet
of the atmosphere can tremendously exceed the adiabatic lapse
rate. This condition causes very great accelerations for any air
particle moving up or down in this layer. The result is that tiny
quantities of air shoot up and down at a great rate. This process
can be seen with the mere eye as shimmering of the air. The very
small particles of air mix readily with the air of the elevations into
which they penetrate, causing a certain uniformity of temperature
in the horizontal direction which makes it possible for this thermodynamically unstable condition to exist for periods of 5 or 10
minutes or more.
On the other hand, a number of the tiniest turbulence elements
in the 6-foot layer flow together, forming somewhat larger tur
bulence elements. The larger ones flow together and so on. Finally,
like a trunk growing out of widespread fine roots, a mass of air
which is warmer than the ambient air emerges from the atmos
pheric layer next to the ground. A thermal is born. It moves up
ward like a balloon, driven by its thermal acceleration and rising
faster and faster and expanding more and more until it dies out at
an altitude where the atmosphere is stable. As the upcurrent
moves, new air flows in at the ground from the sides. This new air
gradually becomes heated, tiny upcurrents flow together until after
a while a new thermal emerges from the same spot. This may take
5 minutes and in other cases an hour or more.
Outside help frequently is required to make the thermal "break
loose" and start upward. A gentle wind, penetrating into the layer
next to the ground, will help to collect the scattered tiny upcurrents
into one sizable one. This is particularly so when the process
takes place on a slope, over which the wind has to climb, thus
starting the updraft in the right direction. A nearby cold spot with
air that has a tendency to flow under the warm air will release
upcurrents. Soaring pilots who have flown over the Texas plains
have always observed reliable upcurrents over the sandy beaches
of the rivers, where undoubtedly the air over the relatively cold
water assisted in causing thermals as much as the actual heating
over the sand and mudflats. There is evidence that a glider pilot
himself can help to release a thermal by maneuvering his ship
over a place that is likely to produce an upcurrent. By stirring the
air at a low altitude, a thermal, which is in formation, can be
made to rise prematurely at the moment when it is needed to avoid
an involuntary landing.
One major factor that spoils the formation of thermals is a high
wind. It disrupts the process of collecting a sufficient number of

little heated particles in the layer next to the ground into a sizable
current. Even if a thermal upcurrent should emerge from this layer,
it soon would be disrupted again by the turbulence of the high
wind. Experience has shown that we have to rely upon other kinds
of upcurrents when high wind velocities exist.
No hard and fast rules can be given as to where thermals are
forming. Too much depends on the physics of the ground in rela
tion to the surroundings, on the position of the sun, the wind
direction and velocity. Many pilots depend on plowed fields,
others favor wheat fields and beaches and a German pilot has made
successful soaring flights with the help of a map of the ground
tooo
FIG. 4. TRAJECTORIES OF THE AIR OVER FLAT COUNTRY

Early in the morning when the atmosphere is stable the air flows almost
parallel to the ground. In midday thermals and turbulence cause irregular cur
rents.
The traces are the paths of floating balloons released between 5:00 and
7:00 A.M. and at 11:30 A.M.
water levels. Thermals, however, are formed by the combined

action of various influences that repeat themselves. Once the search
for thermals at a particular location has been successful, chances
are that there will be an inexhaustible supply of upcurrents that
can be utilized whenever the weather factors are essentially the
same.
The size of thermals, too, varies with many factors. In order to
utilize a thermal, it has to have a size that permits maneuvering a
glider in it. Thermals are largest on calm days and when there is
little chance for their release. On windy days, and under conditions
that favor frequent release of updrafts, the currents are often so

broken up that they can be recognized with the variometer but not
utilized for soaring.
Where there are upcurrents there must be downcurrents. Luckily
the general tendency is for upcurrents to spread out when they
arrive at the higher stable layers. That distributes the downward
motion over a larger space, resulting in gentle downcurrents over
large areas. In contrast, most upcurrents have a comparatively high
speed and are distributed over a small area. The glider pilot stays
in these narrow fields by spiraling tightly.
FIG.
MOTION OF AN AIR PARTICLE OVER FLAT COUNTRY

ON A DAY WITH GOOD THERMALS
Note how the upcurrent is replaced by a downdraft in the same spot over the
airport.
It is important to know that thermals seldom are chimneys of

air going up constantly; rather, they are bubbles repeating them
selves at intervals. Between two upcurrent bubbles there may be
downcurrents at the same spot. Compare Fig. 5, which is the path
of a floating balloon. It shows a thermal upcurrent changed into
a downdraft within a period of a few minutes.
Thermals do not depend on high temperatures as such. What is
necessary is that the ground has a higher temperature than the
air. This is more frequently the case in spring and summer than
in fall and winter. But there are very many occasions during the
latter two seasons when thermals are plentiful. Clear days, on which
the sun's radiation is most powerful, are often associated with cold
waves. Cold polar air acquires a steep lapse rate before it reaches
SOARING METEOROLOGY
125
our latitudes. The ground is still warmer than the air. Both these
factors favor the formation of thermals.
Evening Thermals
Similarly the occurrence of thermals is not restricted to daytime.

During the day heat is being accumulated by the ground. Locations
of high heat capacity, such as water, forests, swamps, and also cities
store up great quantities of heat, even though this may not express
itself in a great rise of temperature during the day. After sundown
heat is sent out to space. As was shown above, the amount of radia
tion is proportional to the fourth power of temperature. Therefore,
places that get very hot during the day cool very rapidly after sun
down. Forests and similar places radiate more slowly and out of
a vast storage of heat. As a result they now become the places over
which thermals form. They are particularly active if they are ele
vated, for the nightly cooling effect on the air makes the lowest
atmospheric layers heavy and sluggish. If this cold air can accumu
late in valleys, thus allowing upper air with a lapse rate that is
still adiabatic to flow over the heat reservoirs, large quiet upcurrent
zones are created which last long into the night. Towns and cities
should always be a source of good thermal upcurrents because
their structure offers comparatively large surfaces to the sun, which
results in the accumulation of much heat. Experience has shown,
however, that the daytime upcurrents of cities and towns are often
so broken up that they are useless to soaring. It is only in the
evening that the turbulence diminishes and allows a large upcurrent
field to form over the city.
Perhaps the most outstanding example of evening thermals is the
upcurrent field that forms with a north wind over the city of
Elmira and the wooded slopes of South Mountain near Elmira. It
is here that thermals were first discovered by the famous pilot,
Wolf Hirth.
Slope Currents
When a wind is referred to as having a certain velocity, the
assumption is that what is meant is the mean velocity of the hori
zontal component of the wind over a certain period of time. For
the wind does not flow in a steady stream parallel to the earth's
surface; in addition to the horizontal flow there is a multitude of
turbulent motions in the air. A gust usually is caused by an eddy
component adding itself to the wind velocity; a lull is composed
of the horizontal wind with the eddy motion going the other way.
The wind varies all the time both in direction and velocity.

If the wind finds an obstacle in its way, like a mountain, it flows
around and over it. \Ve cannot draw an accurate picture of this
flow, because it is just as little stationary as the wind that arrives
at the obstacle. To depict the outstanding characteristics of the flow,
we draw a picture as it would be, were the wind a non-turbu
lent medium. The glider pilot should always keep in mind that
such presentations are idealized and only partly true. The flow
over and around an obstacle changes abruptly, creating different
currents from the ones shown in the pitcure. Fig. 6 gives an idea
FIG. 6. AIR FLOW OVER A MOUNTAIN

The paths of two floating balloons, released simultaneously at the same
spot, do not coincide. In lee one balloon indicates that a downcurrent exists.
Five minutes later the other balloon has to float around a newly formed eddy.
of this. The graph shows the movements of two floating balloons,

which were released simultaneously from an airplane and observed
independently by two theodolite crews. Already during the first
5 minutes they had begun to follow different trajectories. One bal
loon reached the highest point after 14 minutes, the other one
after 18. On the lee side of the mountain, one balloon indicated a
smooth and fast downcurrent. The other one obviously traveled
around an eddy that had formed in lee of the mountain.
Lifting air over an obstacle requires energy which can come only
out of the wind energy. It can be shown, however, that there is
not a sufficient amount of it available. That is, the air tends to
go around the obstacle rather than over it and the wind velocity in
front of the obstacle is reduced, especially in the lower layers. In
the case of an extended mountainous area, this may cause very
low wind velocities in the valleys. Only if a ridge is very wide com
pared to its height can we count on a flow of most of the air over
SOARING METEOROLOGY
127
the ridge. That is why comparatively low but extended slopes, like
dunes, have proven to be more favorable for soaring than individual,
though much higher, mountains. If the ridges lie in the form of
a funnel, air is squeezed into a small space, making the wind veloc
ity rise and thereby increasing the upward component of the
wind. Many of our better soaring sites are so located. A long ridge
rising out of flat country or
water causes the most uniform
and vigorous ridge upcurrents.
The smoother the change
from flat country to the ridge,
the smoother is the air flow. All
steep grades and edges cause tur
bulence and eddies. If a ridge
rises from flat country at an
angle of steeper than 30, it is
FIG. 7. AIR CURRENTS ON THE
likely that a windward eddy WINDWARD SIDE OF A STEEP
forms, inverting the wind direc
RIDGE
tion at the slope. The least disAlmost horizontal flow may change
turbance that must be expected abruptly into almost vertical flow.
is a shift of wind direction be Strong turbulence exists also behind
the edge.
tween valley and summit, caused
by the retarding action of the ridge on the air streaming towards it.
The highest point reached by the air flowing over a ridge usually
is found not over the summit but farther downwind. This is caused
by the inertia of the air and in many cases by lee eddies, which,
so to speak, extend the ridge farther back.
At very steep ridges, quarries, and the like a very turbulent flow
forms. For a time the wind may go almost straight up, at other
moments it flows almost horizontally, as shown in Fig. 7. This
same flow may be encountered also over the edge of wooded areas.
A ridge may have a gentle slope, but suddenly flatten out into a
plateau. Almost inevitably a turbulent zone of slow wind motion
of a thickness of perhaps 200 feet results, above which the flow
continues unabated (see Fig. 8).
Still more attention has to be paid to the lee side of the ridge.
Lee eddies form there, wander off with the wind, and new ones
form. Especially at steep hills, a small but very intense upcurrent
field is created by the lee eddy, as shown in Fig. 9. It is very limited
in size and varies so abruptly that it may act on one wing of the
glider, while the other wing is still in the less disturbed air.
Slope currents are modified by isolation. It has already been
pointed out that thermals are easily released by the upcurrent of

a slope. Thermals should be expected to form at the foot of the
ridge. They go up at an angle, which is given by their rate of rise
and the velocity of wind with which they float. This angle is usually
FIG. 8. THE TURBULENT ZONE OVER A PLATEAU

If a steep slope ends in a flatridge top, the air continues to rise. A shallow
turbulent zone exists over the plateau. The abrupt change of wind velocity
with height makes landings dangerous.
much steeper than the slope of the ridge, so that one has to fly
out from the ridge toward the wind in order to find thermals.
If the sun beats on the slope, much heat is received by the
ground which lies more or less vertical to the sun's rays. In such
cases the ridge upcurrent is in
tensified and the highest point
FIG. 9. EDDY MOTION IN THE

REAR OF THE RIDGE
An intense but narrow and variable
upcurrent field is created by the eddy.
FIG. 10. COMBINED SLOPE CUR

RENT AND THERMAL
When the wind velocity is low and
the sun shines on the lee side of a
ridge, air may move up the ridge
from all sides.
reached by the slope current is not as usual behind the ridge, but
moves to windward over the slope. This was the case when the air
current measurements of Fig. 6 were made.
Insolation may be most intense in back of the ridge. Under slow
winds this results in a combined slope current and thermal. The
SOARING METEOROLOGY
129
air moves up the ridge on both sides, forming upcurrents well to

the rear of the summit. Fig. 10 shows a schematic picture of this flow.
Fig. i i is an actual measurement.
58OO ft
FIG. 11. PATH OF AN AIR PARTICLE IN THE LEE OF A RIDGE

A thermal has been released in lee of the ridge whose upward component
exceeds the downdraft behind the ridge.
Humidity and Condensation

In the atmosphere all gases of which the atmosphere is com
posed, with the exception of water vapor, are found only in their
gaseous state. But it is known that they can be forced into their
liquid state if the temperature is lowered enough and if enough
pressure is exerted.
Water vapor is exactly like the other gases, except that its critical
temperature and pressure are well within the limits of the atmos
phere. As long as it is a gas, it acts just like the other gases of the
air. It is invisible and otherwise unnoticeable to our senses. Its
presence is indicated only by special instruments, the hygrometers.
It must be realized that clouds and fog do not consist of water
vapor. They are formed of small drops of liquid water or small
ice crystals.
As long as the water vapor remains a gas, it does not affect the
thermodynamic processes in the air that were outlined above. But
there are certain conditions under which part of the water vapor
changes to water or ice. We notice this in the appearance of dew
or frost, fog and clouds, and precipitation.
To understand the evaporation and condensation processes, we
assume first that the atmosphere is entirely dry, that is, that it holds
no water vapor whatsoever. The oceans and lakes contain liquid
water; so do many places on land, particularly forests. This water

wants to escape in gaseous form. The pressure with which the water
wants to emerge as a gas from its reservoir of liquid is small. It is
called the saturation water vapor pressure and depends on the
temperature. If the water is warm, its vapor pressure is high. If
the water temperature is lower, the vapor pressure is lower, but
even ice has a certain vapor pressure with which it wants to escape
as a gas. This is impressively demonstrated by the gradual dis
appearance of snow and ice from the ground, though the tempera
ture may stay below the freezing point. Fig 12 shows how the
saturation vapor pressure varies with temperature.
From the oceans, and so on, water vapor now escapes as a gas
into the air which is assumed to have the same temperature as the
water. The water vapor mixes with the other components and is
carried into higher layers by turbulent exchange. The air begins
to contain measurable quantities of water vapor, as shown by a
hygrometer reading.
As the evaporation from the water surface or from forest or the
like continues, the air increases its vapor content until the partial
gas pressure of water vapor equals that at the water's surface. At
that point emanation of water
vapor from the surface natu
rally must stop. The air is sat
urated. Its humidity is 100 per
cent. We say the "relative hu
midity" is 100 per cent.
As Fig. 12 shows, the satura
tion pressure varies with tem
-20-C
-10perature.
or to'
^
_
,.,
We now imagine the saturated
FIG. 12.
SATURATION WATER
.
VAPOR PRESSURE As A FUNCTION air to move over dr7 land; the

OF TEMPERATURE
humidity remains unchanged so
long as the temperature remains
the same. However, if heating takes place, then saturation or 100
per cent relative humidity would correspond to a higher vapor
pressure than is actually there. So, we now have only a certain
percentage of the full possible quota of water vapor. The relative
humidity has become less than 100 per cent.
On the other hand we may have cooling instead of heating. That
means our air would hold more water vapor than its full possible
quota. We get more than 100 per cent relative humidity or supersaturation. That is an unstable condition which cannot last, if the
excess vapor gets a chance to change back to liquid water. It will do
this by falling out on whatever is available. Usually, enough small
SOARING METEOROLOGY
131
particles of hygroscopic dust and salt are contained in the air to

serve as condensation nuclei. Around them the excess water collects
in the form of tiny drops; that is, we get fog or clouds.
The humidity of the air up to great heights is also measured
daily by aerological ascents. It varies greatly from day to day and
in different layers of the atmosphere. Air masses that travel over
maritime regions pick up plenty of moisture, particularly when they
are warm. Typically moist is the air that reaches North America
after sweeping over the Gulf of Mexico. On the other hand, conti
nental air masses are generally dry, particularly when they come
from polar regions where there is less evaporation from the ground
on account of the low temperatures.
The humidity of the air can be expressed by various units of
humidity. Most commonly it is referred to as "relative humidity";
that is, the ratio between the actual water vapor content to the
largest amount possible at the given temperature. Meteorologists
use also the terms "vapor pressure" (in millibars or in mm. or
inches of mercury), "specific humidity" (in grams of water vapor
per kilogram of air), "absolute humidity" (in grams of water vapor
per cubic meter of air), and "wet bulb temperature." For aviation
purposes, the humidity is often expressed by the "dew point tem
perature." The dew point is that temperature to which the air
would have to be cooled in order to be saturated. In other words,
if the air would be cooled just beyond the dew point, fog or clouds
would form.
The Heat of Condensation and Condensation Adiabatic Lapse Rates
A certain amount of heat is needed in order to change liquid
water into water vapor. To evaporate i gram of water of a temper
ature t requires an amount of heat of 597.83-0.647 t. For example,
when boiling water is evaporated, 533 calories are needed for each
cc. of water. That is about six times as much heat as is required
to heat the same cc. of water from room temperature to the boiling
point! The large heat requirements of evaporation are noticed in
everyday life. When water evaporates from our skin, we feel a
chill, because the heat that is used for the evaporation is taken
out of the skin.
When water vapor changes back to liquid water, the same amount
of heat is liberated again. Assume that a quantity of air is cooled
below its dew point; condensation begins to take place and the
corresponding heat is supplied to the air. As a result, the cooling
process becomes retarded. However, the air temperature will not

actually rise because in this case the droplets would evaporate
again immediately.
The amount of water vapor contained in the air when saturation
exists depends on the temperature of the air, as shown in Fig. 12.
Therefore, if condensation takes place at high temperatures, more
water condenses than at low temperatures. The amount of heat of
condensation liberated must therefore be higher at higher tem
peratures or, in other words, the retarding effect on a certain cooling
rate of the air becomes larger at high temperatures.
The temperature in upcurrents drops according to the adiabatic
lapse rate; that is, i C. for every 100 meters of ascent. During
this process the temperature may fall below the dew point. At that
moment condensation sets in and the rate at which the air cools Is
retarded; more so at higher temperatures. The adiabatic lapse rate
changes into another, smaller one. This new lapse rate is called
the condensation adiabatic lapse rate (also called moist adiabatic
or pseudo-adiabatic lapse rate). Near the ground the condensation
adiabatic lapse rate is:
30
86
0.37
20
68
0.44
10
50
0.54
o
32
0.62
jo
14
0.75
20
4
0.86
30 C.
22 F.
0.91 C. per
100 meters.
It varies somewhat with altitude, but the processes involved
are too complex to be explained here. Meteorologists use "adiabatic
charts" which show in graphical form not only the adiabatic lapse
rate, but also the condensation adiabatic lapse rates for all tem
peratures and pressures.
Cumulus Clouds
Cumulus clouds are the billowy, often mountainous types of good
weather clouds that form on days on which there is enough moisture
in the air to lead to condensation when thermal or slope currents
ascend. At first the upcurrent follows the dry adiabatic rate until
condensation begins. Neglecting the expansion of the water vapor
as it is carried upward, we can say that the first cloud droplets form
when the dew point, as measured on the ground, is reached by the
cooling process in the upcurrent. For all practical purposes, we
compute the altitude of the cloud base as the difference between
temperature and dew point temperature in Centigrade degrees mul
tiplied by 100. For example, if the temperature is 25 C., the dew
point 10 C., then the cloud base is found at (25 10) X 1O
1500 meters or about 1650 feet. Taking into account the fall in
dew point owing to expansion of rising air, the cloud base is roughly
SOARING METEOROLOGY
133
at as many thousand meters as one eighth the difference between

air temperature and dew point in C. degrees, or at as many thou
sand feet as one fourth the difference in F. degrees.
Formation of cumulus clouds has two distinct advantages. First,
the upcurrents are now crowned by clouds; that is, they are almost
visible and can be found easily. Secondly, the original upcurrent is
greatly intensified by the heat of condensation. The above table
shows that the condensation adiabatic lapse rate is less than the
dry adiabatic. That implies that cloud upcurrents can easily pene-
Photo by Prof. Alexander McAdie, Blue Hill Observatory,

Harvard University
FIG. 13. CUMULUS CLOUDS

trate into and even continue to receive additional lift from stable
layers in the atmosphere. Meteorologists talk about "conditional in
stability," a condition where the actual measured lapse rate of the
atmosphere is steeper than the condensation adiabatic rate. Such a
condition may exist over a large vertical extent. If a thermal or a
slope current ascends into such a layer, it puts all the available
energy to work and tremendous cumulus clouds with very great
vertical wind velocities form. They are called cumulonimbus. Up
currents of over 50 feet per second have been found in such clouds.
Frequently electric disturbances are connected with such overgrown
cumuli; we then have local thunderstorms. In 1939 the American
altitude record was more than doubled by a flight into such a
thunderstorm.
The meteorologist can compute the size of upcurrents in cumulus
clouds from the temperature and humidity distribution of the
atmosphere, and measurements by glider pilots have verified such

computations. With some experience the pilot himself can readily
form an opinion on the intensity of the vertical currents in a
cumulus cloud. The bulging edges of the cumuli give an impressive
picture of the mighty forces at work. But the air movement at the
edges is greatly reduced by the friction between the ascending mass
of air in the cloud and the free atmosphere. Inside, the vertical
speeds are even larger, though it is reported that the motion is
more uniform compared to the turbulent boundary.
Some ten years ago much research by Peter Riedel and the author,
then with the German Research Institute for Soaring, was spent on
F. Ellerman, Mt. Wilson Observatory
FIG. 14. CUMULONIMBUS ("THUNDERHEAD")

finding out if there is any regularity in the upcurrent field of a
cumulus cloud. For this purpose hundreds of systematic flights were
carried out. A small airplane with recording meteorological instru
ments attached under its wings was taken under a cumulus, then
the engine was stopped entirely and the ship glided down, circling
under the cloud. The barograph recordings of the plane allowed
exact determination of the vertical currents encountered, as Fig. 15
shows. Ground crews measured the motion of the cloud and the
position of the plane in regard to the cloud. It could not be estab
lished that any particular part of the cumulus cloud would regu
larly furnish better lift than other parts. Still, a number of soaring
pilots maintain that better lifts are obtained on the windward side
of cumuli and on the side exposed to the sun.
More important to practical soaring are the experimental results
SOARING METEOROLOGY
135
concerning the variation of upcurrent intensity with time. Like a

thermal, a cumulus cloud is not stationary. It forms, builds up, de
creases and disappears. The life cycle of a cumulus cloud varies from
a few minutes to several hours. It can be determined easily in each
case by observation of the clouds. When a cumulus is created, as
described above, very small droplets form first. They are suspended
in the air, having only a very low rate of fall. But all the time small
drops grow and coalesce, forming bigger ones. These fall faster. If
their rate of fall exceeds the upcurrent, they might even come to
5000
Feet
4000
3000
Up
Drail
10
2.000
1 i
Ft/sec.
s
0+
1000
1 t
i
Draff
0
Minutes
IS
FIG. 15. UPCLRRENTS UNDER A CUMULUS CLOUD

The normal sinking speed of the airplane is shown by the steep line from
left to center bottom. The barograph trace shows the actual loss of altitude
under the cumulus. The arrows are the upcurrents as computed from the differ
ence of actual and normal sinking speed of the plane.
the ground in the form of a shower. Though this is not usually the
case with ordinary cumuli, comparatively large amounts of water
collect in large drops in the lower part of the cloud.
The original thermal bubble shoots up through its adiabatic state,
then through its condensation adiabatic state; it finally reaches a
region of the atmosphere, perhaps an inversion, where the upcur
rent particles become heavier than the surroundings. On account of
inertia, the advance part of the upcurrent pushes deeply into the
inversion. After the energy of movement is spent, the particles tend
to sink again. But they meet the rest of the upcurrent which still
tends to go up. So the currents spread out sideways. If this process
is pronounced, the cloud takes on the shape of a huge anvil.
After all of the bubble has bumped into the inversion, the entire
mass of air swings back to earth. It is now distributed over a larger
space so that the downcurrent is slow. Nevertheless, the droplets in

the upper part of the cloud evaporate at a fair rate as they are
carried into lower warmer regions. It takes some time for the
downcurrent to proceed (o the lower part of the cloud. In this
region, as shown above, there is an accumulation of large drops,
which it takes some time to evaporate. This shows that the cloud
remains in existence for some time while downcurrents are active.
Cloud Streets and Waves
The type of cumulus clouds best liked for soaring is the one just
described. Its formation is caused by thermals and occasionally by
slope currents. It occurs on calm days and when moderate winds
prevail. On stormy days the cumulus takes on another shape.
High winds are very turbulent because all the obstacles on the
ground over which the wind passes cause innumerable eddies and
irregular currents to form, which float along with the wind, only
slowly diminishing under the influence of friction and possible
thermal forces. The continuous action of all the up- and downcurrents causes a general adiabatic lapse rate to be established in
the turbulence zone. Compare Fig. 3. The mean temperature of
the layer over which the turbulence spreads remains the same.
If there was originally a stable lapse rate or an inversion, we get
now a temperature rearrangement so that the temperature rises at
the ground and decreases at the upper end of the turbulence zone
with the same mean temperature of the layer. In that way an in
version, the "turbulence inversion," is formed between the turbu
lence zone and the upper atmosphere. If heating takes place at the
ground, the temperature of the entire turbulence zone with its
adiabatic lapse rate goes up. Therefore the turbulence inversion
becomes smaller and may even disappear.
If the moisture content in the turbulence zone is high enough,
condensation takes place and cumulus clouds form. In spite of
their heat of condensation, however, they cannot shoot up freely
because of the turbulence inversion. In contrast to the "heat cu
muli," the "turbulence cumuli" are flat; the more so, the more
pronounced is the turbulence inversion. In contrast to the "heat
cumuli," the "turbulence cumuli" have a pronounced upcurrent
region to the windward and downcurrents, downwind, where they
dissolve.
Turbulence cumuli are no good for altitude soaring. But it hap
pens frequently that a certain regularity in the turbulence causes
the turbulence cumuli to arrange themselves in a quasi-regular
pattern. That greatly facilitates distance soaring along chains of
such clouds.

Not infrequently turbulence develops into a regular rolling mo
tion, particularly in the lee of long-stretched mountain ranges, from
which large lee eddies float off. Then we get long-stretched cloud
banks of stratocumulus arranged all over the sky in long bands and
ribbons. They may be arranged perpendicular to the wind direc
tion, but are also known to form parallel with the wind. Upcurrents
are found under the clouds, downcurrents in the open spaces. This
is a favorable condition for long-distance soaring.
Turbulence cumuli are not bound to a turbulence zone near the
ground. In fact they may form on top of a ground inversion, if a
strong wind blows over the layer of air next to the surface. If the
FIG. 16. ALTOCUMULI

wind blows over water or sand, waves are formed. Similarly, air
waves form in the ground inversion layer when the upper air flows
over it. These waves are much larger in size than the water and
sand waves and apt to topple over. Thus turbulence starts and
penetrates both downward into the inversion and upward into the
layer of high wind. If there is sufficient moisture, turbulence cumuli
come into existence.
The same process of wave formation is what causes the mackerel
sky; that is, high clouds of the cumulus type (see Fig. 16). Alto
cumuli are most common at altitudes of from 10,000 to 15,000 feet.
They form from air waves caused by wind shifts at inversions at
these elevations. Since such clouds are not subject to diurnal varia
tions, they might become an important source of upcurrents for
duration flights. German attempts to soar under altocumulus clouds
after airplane tow were promising.

Polar Front and Air Masses
On previous pages occasional reference was made to "air masses."
The "air mass analysis" for weather forecasting is now widely em
ployed in this country and familiarity with its concepts is most
useful to the glider pilot. However, the subject is too broad to be
treated fully here. Only a general outline can be given with some
emphasis on the cold front which is particularly important for
soaring.*
If the same train of thought that led to the explanations of
thermals is applied to the atmosphere as a whole, we would con
clude that permanent upcurrents exist over the equatorial regions,
where the heating by the sun is most intense. Then downcurrents
should be the rule over the cold polar regions. A general circulation
would result in such a manner that the northern hemisphere would
have prevailing north winds near the ground and south winds
aloft. This simple scheme, however, is spoiled for two reasons. First,
because of the presence of continents and oceans. In summer the
land is warm compared to the water. A circulation from water to
land in the lower atmospheric layers and a flow from land to water
at greater heights results. In winter this process works the other
way around. The trade winds are the result of this action between
land and water.
Secondly, there is the rotation of the earth. When air leaves the
North Polar region, traveling south toward the equator, it has
little east-west or west-east velocity and tends to continue straight
south. The earth rotates from west to east and the velocity of the
earth's surface increases as the distance from the axis increases;
that is, the farther away we go from the pole. The result is that
the earth, so to speak, slips away under the air towards the east.
We notice this relative motion as an east wind.
Since the circumference of the earth increases as we go south
from the North Pole the air from the small polar region distributes
over a larger and larger area as it proceeds south, so that its south
ward component of movement gradually diminishes. At a latitude
of about 65 nothing is left but an east wind blowing around the
earth.
It can be shown similarly that the air which starts north from
the equatorial regions is gradually bent over into a west wind. This
current covers our latitudes, hence our prevailing westerly winds.
At a latitude of about 65 we now have a cold easterly current
* For more detailed information, the book on Aeronautical Meteorology by
George F. Taylor is recommended,

on one side and a warm westerly current on the other side. The
demarcation line is called the polar front.
What happens is depicted in Fig. 17. Waves form. On account
of the small density difference between the polar and the equatorial
air, the waves are of tremendous size, extending over hundreds of
miles.
Fig. 17 shows how a wave, formed at the polar front, changes
over a period of time. \Varm air penetrates the cold region and in
back cold air sweeps around into the originally warm region. But
the relative movements of the warm air and cold air are more
Warm
Cold
"Cold
Warm
Warm
FIG. 17. WAVES IN THE POLAR FRONT
complicated than that. The air moves not only in the horizontal,
but also in the vertical direction; the warmer and therefore lighter
air slides up over colder air, which sweeps around, and pushes
under the warm air.
Fig. 18 shows one phase of such a wave in greater detail. The
lower part of the drawing corresponds to the presentation in the
previous figure. The upper part is a vertical cross section through
the line AB.
On the ground we find the warm air only in the triangular
southern section, the "warm sector." The line where the warm
air begins to slide from the surface up over the cold air is the warm
front. The line which separates cold and warm air where cold air
advances is called a cold front.
Along the warm front surface, the warm and relatively moist air
moves slowly upward, too slowly to furnish upcurrents for soaring.
As it comes under lower pressure, it cools. More cooling is caused
by a certain amount of mixing between the cold and warm air

along the warm front surface. Consequently, extended cloudiness
is caused, beginning with cirrostratus at great heights down to
stratus and fog. Long stretches of steady rain accompany the pas
sage of the warm front surface. There is good warm weather in
the warm sector, but the northwest current that brings the polar
air around behind the cold front lets the temperature drop abruptly
when it passes.
The whole phenomenon, as presented in Fig. 18, is idealized, but
can often enough be recognized clearly on the weather map. If
Warm
FIG. 18. DISTRIBUTION OF WIND, TEMPERATURE AND CLOUDINESS
IN A "Low"
At the warm front warmer air slides up on top of colder air. At the cold front
polar air pushes under warm air.
isobars are drawn, it is found that we have the detailed structure

of a "low." The wave moves away from the place of its origin and
travels along in our prevailing westerly current. All the time its
own development, the play between colder and warmer air, con
tinues.
The Cold Front
A cold front can be hundreds, even thousands, of miles long. It
stretches out in a sweeping line, traveling from west to east at a
speed the order of magnitude of which is 25 to 40 miles per hour.
The heavy cold air behind it drives the warmer air in front out of
the way and upward, much as a wedge would that is driven under it.
If the warm air is particularly warm, that is, light in places, and if

it is quite unstable, less energy is required to remove it. In this
case the cold air advances faster and the front bulges out. On the
other hand, the front may get retarded. If there is a mountain
range in the way, the cold air is held back until enough has accu
mulated to flow over the range. By this time much of the cold air
has advanced around the mountains. The gap in the front may
close again behind the obstacle before the cold air reaches as high
as the mountain tops. Naturally the sudden lifting process of the
front is thereby greatly reduced and broken up within the moun
tain region.
Fig. 19 shows the schematic picture of a front passage, based on
actual measurements. The lower part of the drawing contains a
15000
Feet
10000
5000
FIG. 19. THE PASSAGE OF A FRONTAL THUNDERSTORM

(COMPARE TEXT)
reproduction of the trace of a wind-registering instrument. Reading
from right to left, there was a 10 m.p.h. SSW wind, the air move
ment in the warm sector. Compare Fig. 18. Very shortly before the
arrival of the front, the wind died down. The calm before the
storm! Then, as the cold air broke in, the wind on the ground
shifted over west to northwest and jumped up to a maximum of
41 m.p.h. Had there been a temperature registration, it would have
shown a temperature drop of many degrees during this short period.
In the upper part of Fig. 19 the warm air is indicated toward the
right, the cold air toward the left. The actual boundary line be
tween the two is doubtful. It cannot easily be measured because
nobody wants to fly into such regions for the sake of scientific

measurements. And even then the measurements would be difficult
to interpret because of the adiabatic and condensation adiabatic
changes taking place when the vertical movements get under way.
From experiments with models and from the measurements of air
currents that were occasionally made by unlucky glider pilots who
happened to fly into this region, we believe that the picture corre
sponds closely to the actual conditions. Where the "head" of the
cold air rushes in, a tremendous eddy, often a regular roll, forms.
That accounts for the gust which accompanies the passage of a
front. In front of this roll and over it the warm air is thrown
upward. Condensation takes place, a tremendous cumulonimbus
builds up all along the line of the front. It destroys stable condi
tions and inversions, though the altitudes at which inversions used
to be are often still indicated by the fanning out of the thunder
cloud at various heights.
The upcurrent field starts away out in the warm air. As the
upward movement continues, heavy black clouds form at that
place; that is, the front moves on. The clouds are the result of
condensation. As shown before, the heat of condensation increases
the upcurrents. The highest vertical velocities and the greatest tur
bulence are found in the forward part of the clouds. The upcurrents
sustain the clouddrops. They also sustain the raindrops, which form
by growth and coalescence from clouddrops, for even the largest
raindrops fall only with a velocity of about 20 feet per second. At
some altitude the temperature goes below freezing; ice crystals form.
They collect cloud droplets around themselves, creating pellets com
posed of ice and water.
These may fall for a while. But if a current moves them up below
the freezing point again, they coat with a layer of ice, and we get
hail. All these raindrops, pellets, a^nd hail are shoved back by the
big eddy current. When they arrive in the region of the downward
branch of the eddy, they fall through fast because the downcurrent
increases their own rate of fall. The precipitation arrives on the
ground shortly after the big gust has passed.
After the period of the precipitation, the front has passed. We
are now in the cold air mass, which gradually increases in thick
ness. As it flows over the ground, which is still warm and wet, the
lower layers are heated and supplied with moisture. Cumulus clouds
form and soaring conditions become excellent, to the consolation
of those pilots who have missed connecting with the upcurrent
field in the warm air just ahead of the front.
There is only one rule to front soaring, but it is all-important:
always stay in front of the big thundercloud and never allow it
SOARING METEOROLOGY
143
to catch up with you. The inside of a frontal thunderstorm is, to

say the least, unpleasant.
The Moazagotl Condition
In recent years the attention of both glider pilots and meteorolo
gists has been focused on a condition which permits soaring to very
great heights, more than 20,000 feet, with comparative ease. It is
now known that this condition can exist at various places all over
the world. It was first discovered in the Sudeten Mountains in
Germany. Many generations ago the inhabitants began to observe
that occasionally a large, isolated cirrus cloud formed at great
heights. This cloud proved to be the forerunner of bad weather
and is used by the natives as a weather indicator. They named it
"Moazagotl," which is Silesian dialect and may be freely translated
into something like "foe's beard."
The weatherwise members of a glider school located in this
region paid due attention to the Moazagotl cloud, and soon found
that soaring conditions were greatly improved when it was in
existence. As the experiences of many soaring flights were pieced
together, the following picture evolved. With a wind velocity of
about 40 m.p.h. the usual lee eddy behind a mountain range
stretching out perpendicular to the wind is not formed. Instead,
there is a continuous strong downcurrent. But 5 miles farther back,
a strong upcurrent field exists. Behind the upcurrent field, downcurrents follow, then a new upcurrent field, 5 miles distant from
the first, a third region of downcurrents and a third region of
upcurrents, again about 5 miles farther back and all parallel with
the mountain range. The fields of up- and downcurrents do not
move with the wind, but are stationary. In the atmosphere near the
ground, remarkable wind shifts occur. In some places the wind
direction is entirely opposite to the prevailing wind. At great
heights the huge cirrus cloud, consisting of ice crystals, caps the
entire region.
Since the glider school is located under the third upcurrent re
gion, flights were at first made there. Later it was found that even
better conditions existed in the second field, and in September 1937
a flight into the first region and deeply into the Moazagotl cloud
led to the then record altitude of 22,000 feet.
After an exhaustive study of the condition, the following ex
planation of the phenomenon has been offered. It is illustrated by
Fig. 20.
If the wreather situation causes a flow of air which is cold in the
layers next to the ground and if a mountain range is in the path of

this air, then the cold air will collect on the windward side of the
range until it reaches high enough to flow through the mountain
passes and finally over the entire ridge. On the lee side, however,
the cold air flows off quickly into the center of the low. There will
develop a difference in level of the cold air. It is high in front of
the range, low behind it. The warmer air current aloft approaches
with great velocity and, falling down the precipice of cold air,
forms large standing waves which extend through the entire atmos
phere up to the stratosphere. If the warm air is sufficiently un
stable and moist, the amplitude of the waves increases with height
and a Moazagotl cloud forms. The waves have a length of from
4 to 6 miles; the amplitude in the lower layers is only about 1500
FIG. 20. THE AIR CURRENTS OF THE MOAZAGOTL CONDITION

Near the ground there are stationary rolls; aloft standing waves of amplitude
that increase with height are capped by a cirrus cloud.
feet. At times the waves may work their way down to the ground,
but usually a number of large "rolls" of air form near the ground,
filling out the space under the crests of the waves. The diameter
of these rolls is approximately the same as the height of the moun
tain range, and the upward branches have also a sufficient vertical
velocity for soaring. The Moazagotl condition is more frequent
during the colder seasons of the year.
Such conditions were first encountered and used in America by
John Robinson over Sun Valley, Idaho, in 1940. The records of the
observatory on Mt. Weather in Virginia contain observations of
Moazagotl clouds behind the Blue Ridge Mountains, and the author
has observed the same phenomenon from Mt. Washington Ob
servatory in New Hampshire. It will require some patient experi
menting to locale exactly the upcunent fields and to pick the
collect weather situation. But there is a decided opportunity for
exceeding the American, if not the international, altitude record.
CHAPTER VII
INSTRUMENTS
By Charles H. Colvin
and August Raspet
kLTHOUGH IT is BEST to give primary instructions in a

glider unequipped with any instruments, several are useful for
soaring. For maximum performance of a modern sailplane at least
six are necessary. These are in order of their importance: vari
ometer, altimeter, airspeed indicator, turn and bank indicator,
compass and clock. Some pilots may argue the relative importance
of the turn and bank, the airspeed, and the altimeter, but all will
agree that the most essential is the variometer. However, the first
used in gliding is the airspeed indicator so it will be described first.
AIRSPEED INDICATOR
The airspeed indicator shows the rate at which the glider is
moving through the air. It is a differential pressure gauge which
is actuated by the pressure difference generated in a pitot-static or
pitot-venturi tube.
FIG. 2
FIG. i
AIRSPEED INDICATOR
Two tube lines connect the instrument to the pressure tube,

which is mounted in a position where it receives an undisturbed
flow of air.
The indication of the airspeed indicator depends upon the
145

density and temperature, as well as upon the speed of the air. It is
graduated in units of speed for "standard" atmospheric conditions;
that is, barometric pressure of 29.92 in. of mercury (1013 millibars)
and a temperature of 15 C. For other pressures and temperatures
corrections must be applied.
The airspeed indicator is shown in Fig. i, in which the inside
of the diaphragm is connected to the pitot tube, and the case
(outside of the diaphragm) is connected to the static or venturi
lube. Increase of air speed causes increase in the pressure differ
ence and expansion of the diaphragm. This lifts the wire bridge
Kollsman Instrument Company, Inc.

FIG. 3. PITOT-STATIC TUBES FOR AIR SPEED
and the arm which rests upon it. The movement of the arm turns
the rocking-shaft and its arm which presses against the sector arm.
The latter turns the sector, which engages the pinion, turning it
and the hand. The hairspring keeps all parts in contact with each
other.
The face of the airspeed indicator is seen in Fig. 2. A typical
pitot-static tube is illustrated in Fig. 3.
ALTIMETER
The altimeter shows the altitude of the glider in relation to sea
level or to the earth. It is an absolute pressure gauge the dial of
which is marked in units of altitude.
The altimeter operates from the difference in pressure inside and
outside an airtight diaphragm. The pressure outside the diaphragm
is that of the inside of the instrument's case, which is airtight and
is connected to the static tube of a pitot-static tube. This connection
is generally joined to the static connection of the airspeed indicator.
The pressure on the diaphragm is therefore the same as that of the
outside air at the static opening of the pitot-static tube.
Altimeter readings are correct only under "standard" conditions;
that is, when the barometric pressure at sea level is 29.92 inches of
mercury, the temperature at sea level is 15 C. (59 F.), the tem
perature gradient is 1.98 C. (3.56 F.) per 1000 feet, and the
barometric scale of the altimeter is set to 29.92 inches. Under all
other conditions, to obtain the altitude, corrections must be made
for pressure by setting the altimeter barometric scale to the al-
INSTRUMENTS 147

FIG. 4. SENSITIVE ALTIMETER

FIG. 5. SENSITIVE ALTIMETER
!48 FLIGHT WITHOUT POWER

timeter sea level pressure, and for temperature, by the use of tables
or a computer.
The operating and setting mechanisms of the Kollsman Sensitive
Altimeter are shown in Figs. 4 and 5. The diaphragm D expands
as the outside pressure is reduced, moving a link l^ 1 and arm Al and
thus turning rocking-shaft R. The weight of the diaphragm is bal
anced by balance arm B which is supported by spring E and is
connected to the rocking-shaft by link L2 and arm A2. Rockingshaft R carries sector S, which drives pin
ion Px on shaft K1? which also carries gear
G2, and hairspring C. Gear Gj drives pin
ion Po on shaft K 2, which also carries
pinion P3 and the long hand Hx.
Referring now to Fig. 5, pinion P3 drives
gear G3 through G2 and pinion P4 . Gear
G3 is on shaft K4, which also carries the
short hand H2 . Dial D L is fixed in the
case, but dial D 2 , together with the mechanism, is turned by turning knob N. This,
IMG 6 KOLLSMAN
through stem M and pinion Qlf turns
SENSITIVE ALTIMETER
gears Fx and F2. Gear Ft turns the mech
anism directly and gear F 2 turns the dial D 2 through pinion Q2 and
gear F3 .
The face of the altimeter is seen in Fig. 6.
BALL BANK INDICATOR
The ball bank indicator is shown in Fig. 156. It consists of a ball
within a curved glass tube. The tube is filled with liquid except
for an air bubble which is left in a standpipe at one end of the
tube (out of sight). When a turn indicator is used the bank indi
cator is usually combined with it. The ball bank is helpful in
keeping the sailplane level laterally in straightaway flight and
essential in making turns of narrow diameter when the sailplane is
steeply banked, especially in cloud flying.
VARIOMETER
The variometer is also known as a climb indicator or a vertical
speed indicator. It shows the rate of change of altitude of the glider
and is operated by the rate of change of atmospheric pressure.
The capillary leak type of variometer as made by Kollsman,
Pioneer, and Askania, comprises a sensitive differential pressure
gauge, an air chamber (usually heat insulated), and a leak tube.
INSTRUMENTS
149
The Kollsman instrument, as illustrated diagrammatically in

Fig. 7, has the insulated chamber (not shown in the picture)
within the instrument case, the mechanism being within the cham
ber. The inside of the diaphragm is connected directly to the out
side air through a static tube (which may be the static tube of the
airspeed indicator pitot-static tube), while the outside of the dia
phragm is subject to the pressure of the air in the chamber, which
is connected to the outside air through a leak tube.
As the glider rises, the pressure inside the diaphragm becomes
less than that in the chamber outside the diaphragm and the dia
phragm moves inward. Amplifying mechanism transmits the move
ment to the hand, causing it to move up.
FIG. 7. VARIOMETER
FIG. 8. KOLLSMAN VERTICAL

SPEED INDICATOR
The face of the instrument is seen in Fig. 8.

The Cobb-Slater variometer employs two tapered tubes, the
"leaks" being variable in accordance with the position of a small
indicating pellet in each tube. This instrument is shown diagram
matically in Fig. gA. The size of the tapered tubes is greatly ex
aggerated in the picture.
The top of the UP tube and the bottom of the DOWN tube are
connected to the outside air, while the bottom of the UP tube and
the top of the DOWN tube are connected to an air chamber. As the
glider rises, the relatively greater pressure in the chamber causes air
to flow out. As the pellet in the DOWN tube is at the bottom, seal
ing the tube, no air can flow through this tube. It therefore goes
through the UP tube, lifting the pellet in this tube until the passage
way around the pellet is sufficient to balance the pressure differ
ence. The height the pellet is lifted is a measure of the rate of
ascent.

Similarly, as the glider goes down the air flows from the outside
to the chamber and as it cannot pass the pellet at the bottom of
the UP tube it lifts the pellet in the DOWN tube until the leakage
past the pellet is sufficient to balance the pressure.
The face of the Cobb-Slater variometer is seen in Fig. gB.
TO
BOTTLE
DOWN
(RED)
TO
STATIC
A. Diagrammatic Sketch
B. Face
FIG. 9. COBB-SLATER VARIOMETER
The Horn Variometer

Like the Cobb-Slater instrument, the Horn variometer was de
veloped specifically for soaring. Likewise, the fundamental principle
of operation is the same as that of the Cobb-Slater and of the
capillary leak variometer: air in an insulated chamber is allowed
to expand or is compressed through a small leak as the aircraft
climbs or sinks. The Horn variometer is exactly the same as the
Cobb-Slater instrument in that the capillary is combined with
the pressure-measuring movement, but differs from the latter in the
method of measuring the small pressure differential across the
capillary leak. The movement of the Horn variometer (Figs. 10
and 11) consists of a very light metal vane which is attached to
a shaft rotating in jeweled bearings. The vane is centered in a cylin
der closed at its ends. A small sector divides the cylinder into two
parts. Into this sector are drilled the orifices through which air
flows into or out of the cylinder. When the sailplane climbs air
expanding in the insulated chamber flows through the orifice A into
the left portion of the closed cylinder of the movement and the
INSTRUMENTS 151
resulting pressure then causes the vane to rotate. The rotation of
the vane is restrained by the centralizing torque of the hairspring.
The angle through which the vane has rotated is thus a measure
of the pressure differential causing the air to flow from the insulated
JEWELED
BEARING
FIG. 10. MOVEMENT OF THE HORN VARIOMETER

chamber. The angular movement of the vane, indicated on a dial
by a needle, corresponds to the rate of climb of the sailplane.
In order to make the indication of the Horn instrument inde
pendent of the attitude of the glider, the movement is carefully
FIG. 11. SCALE AND MOVEMENT ASSEMBLY OF

THE HORN VARIOMETER
balanced with the instrument in different positions. When this
static balancing is completed, the indications of the instrument not
only are unaffected by changes in attitude of the glider but also
are not subject to the centrifugal acceleration of the glider while
spiraling. In contrast, it may be stated that the indications of the

pellet-type variometers are directly affected by centrifugal accelera
tions.
The Electrical Variometer
Although still not available on the market, the electrical vari
ometer merits discussion in this book because its characteristics
particularly adapt it to exploration for thermals. Unlike other
variometers, the electrical instrument does not depend upon a pres
sure differential for its action, but upon the flow of air at a low
pressure differential from or into the insulated air chamber. So
low is the pressure differential that it requires but a short time
BATTERY
TO STATIC
PRESSURE
TO INSULATED
CHAMBER
FIG. 12. ELECTRICAL VARIOMETER

interval to be developed; consequently, the instrument possesses
an inherently short time lag. The lag is, in fact, so small that the
turbulence in the air flow around the glider may cause the instru
ment to fluctuate. In an experimental instrument constructed by
the author and Robert McConnell in 1940 a lag was introduced in
order to reduce the fluctuations caused by turbulence. The im
portance of a short time lag in considering a variometer for soaring
will be discussed in a subsequent section.
The active elements of the electrical variometer are two resistance
coils of fine wire. These coils are placed in a heat insulated cell
through which may flow the air expanding from the insulated air
chamber. The active coils R3 and R4 (Fig. 12) are connected by
two fixed resistances R : and R2 into a circuit known as a Wheatstone bridge. Current from a battery heats the active coils to a high
temperature. The variable resistance R0 is adjusted so that, with
INSTRUMENTS 153
no air flowing through the cell, no current flows through the indicat
ing meter M, which, as a result, assumes its zero center position. The
circuit is now said to be balanced and is ready for operation. When
the glider climbs, air expands in the insulated chamber and flows
through the variometer cell from R4 toward R3 . As the air strikes
the heated coil R4, heat is transferred from that coil to the air,
thereby raising the temperature of the air and lowering that of the
coil R4 . On passing over the coil R3 the air, now quite hot, cannot
cool R3 as much as it did R4 . This unequal cooling of the active coils
causes an unequal change in the resistance of the active coils. The
Wheatstone bridge circuit therefore becomes unbalanced and there
CAPILLARY
TO STATIC
PRESSURE
LEAK
^TO INSULATED
"AIR CHAMBER
U-TUBE
MANOMETER
LIQUID
FIG. 13. CAPILLARY LEAK VARIOMETER USING

LIQUID MANOMETER
flows through the meter a current which is indicated by a deflection
of the needle of the meter. The amount of deflection on the meter
is a measure of the air flow through the variometer cell and, con
sequently, is also a measure of the rate of climb of the glider. If the
air flow is from R3 to R4, the bridge is unbalanced in the opposite
direction, giving a downward deflection to the needle of the meter.
This is as it should be, indicating that the glider is sinking.
The Capillary Leak Variometer Using a Liquid Manometer
The capillary leak variometer employing a liquid manometer for
measuring the pressure differential operates exactly as the capillary

leak variometer employing a diaphragm-pressure indicator except
that the pressure is measured by the change in levels of the liquid
in the arms of the U-tube manometer, Fig. 13. In order to permit
small rates of climb to be indicated, the two arms of U-tube are
sometimes filled with two nonmiscible liquids differing slightly in
density. This difference in density makes possible the indication of
very small pressure differentials. As an example, one may choose
as the two liquids, oil and water, which have densities of 0.8 and
i.o grams per cc, respectively. The readings of such a manometer
would be five times that of a manometer using only water. Such an
instrument is, of course, subject to errors resulting from banking of
the ship. In addition, the indicator is affected by accelerations
during spiraling. In England, David Dent has ingeniously re
moved the errors resulting from banking and pitching of the
glider by making the arms of the manometer in the form of con
centric tubes. This instrument was much used in early thermal
flying in England, but has since been displaced by instruments
with a shorter lag.
Total Energy Variometer

A novel technique for the use of the variometer in soaring has
been offered by Arthur Kantrowitz.* He suggested that the strength
of an upcurrent may be determined more positively from a sailplane
if the effect of zooming or diving is eliminated from the variometer
readings. Many novices in soaring have experienced what is com
monly called a "stick thermal." This effect is merely a zoom which
is interpreted by the pilot as a rising current when actually it is an
erroneous interpretation of the variometer's reading. Kantrowitz
proposed that the total air pressure variation be measured instead
of only that due to barometric pressure change. The conventional
variometer is connected with its static opening joined to the static
side of the pitot-static head or else with the static opening exposed
to the air pressure in the cockpit. In the total variometer connection
the static opening is, however, connected to a specially constructed
venturi tube. This connection subjects the air in the insulated
chamber to a pressure of J/g^v2 -j- @gh. It can be seen that this pres
sure is exactly proportional to the total energy of the glider i/^mv2 -Jmgh. The readings of a variometer so connected will be a measure
of the rate of change of total energy of the sailplane. The rate of
change of total energy of a sailplane is altered only when the sail
plane encounters an upcurrent or a downcurrent. In actual opera
tion the venturi of the total energy variometer subjects the air
* Jour. Aero. Sciences, Oct., 1940, p. 523.
INSTRUMENTS
155
in the insulated chamber to a negative pressure equal to i/g^v2 -[gh. If the glider is zoomed the air speed is reduced and conse
quently the negative pressure on the air chamber is reduced. Air
would therefore tend to flow into the air chamber. However, the
reduction in pressure resulting from the altitude gained in the
zoom would tend to cause the air to flow from the air chamber.
If the venturi is of certain proportions, the tendency of the air to
flow into the chamber is exactly compensated by the tendency to
flow out. Fig. 14 shows a venturi which will accomplish this result.
Besides the elimination of the effects due to kinetic changes, the
total energy variometer offers a distinctive safety feature. Attempts
at stretching the glide have all too often ended in stalls and spins.
JL
3"
TO
STATIC CONNECTION
OF VARIOMETER
FIG. 14. VENTURI FOR THE TOTAL ENERGY VARIOMETER

Because the total energy variometer does not show a small loss of
energy if the air speed is reduced, it will not, if observed during a
stretched glide, give the pilot as encouraging an indication as does
a conventional variometer. If for this safety feature only, the total
energy variometer merits recommendation.
In selecting a variometer for use as a total energy variometer one
should be chosen which has a time lag at least as small as the re
sponse time of the sailplane. The change in pressure resulting from
a change in air speed should coincide in time with that resulting
from the change in altitude. With variometers having a long time
lag, it may happen that the pressure changes do not coincide in
time, causing large errors in the measured effects.
Considerations in the Selection of a Soaring Variometer
In the field of instrumentation there are many devices called
variometers. It is unfortunate that in soaring parlance the rate of
climb indicator has been called a variometeryet considering its
many variations and errors the rate of climb indicator has perhaps
aptly been termed a meter of variations. The characteristics of

variometers and their effect upon the performance of the pilot and
his ship will be discussed in the following paragraphs.
A variometer's usefulness in a sailplane is determined primarily
by its ability to indicate a small rate of climb. Whether that indica
tion be the true rate of climb or whether that indication lag in
time behind the true rate of climb is of secondary importance to
a soaring pilot. The ability of a variometer to detect and indicate
a small rate of climb is called its sensitivity. The sensitivity of
different variometers is best expressed in terms of the rate of
climb corresponding to a certain deflection of the indicator on the
dial or face of the instrument. It is sometimes convenient to esti
mate the sensitivity by determining the rate of climb corresponding
to the smallest graduation in the scale of the variometer. This
method is not, however, a reliable criterion of a variometer since the
graduation spacing varies among the various instruments. A sensi
tivity at least as good as 3 feet per second of climb for an indicator
deflection of one inch is required of a variometer to be used for
soaring flight.
Not only must a soaring variometer be able to indicate small
rates of climb, but it must indicate changes in those rates of climb
with the least possible delay. The delay of the indication, called
the time lag, is an inherent property of all variometers. A certain
interval of time is required for the pressure differential between
the atmosphere and the insulated air chamber to build up to a
value large enough to actuate the mechanism of the variometer.
Obviously, an instrument capable of operating on very small pres
sure differentials will possess a small time lag. Because the indicator
of a variometer lags behind the true rate of climb, the pilot of
a sailplane is actually furnished a history of the rates of climb he
has experienced. During soaring flight a pilot can explore a thermal
more easily if the history of his climb rates is more recent. It is
for this reason that so much emphasis has been placed upon the
development of a variometer possessing no appreciable time lag.
Research has, in fact, been initiated on instruments which, it is
hoped, will predict the occurrence of thermal convection. Cf. Tem
perature Gradiometer.
Another concept, called resolving power, may emphasize better
the importance of a small time lag in a variometer for exploring
thermal structures. The resolving power of the combination of a
sailplane and a variometer is a measure of the smallest thermal
which can be detected by a pilot of such a sailplane while flying
at normal cruising speed. A measure of the resolving power is the
product of the cruising speed of the glider in feet per second and
INSTRUMENTS
157
the time lag in seconds of the variometer. Consequently, to retain

a fair resolving power in a modern high-speed sailplane, the sail
plane must be equipped with a variometer having a small time lag.
Of all the instruments on the instrument panel of a sailplane the
variometer is probably observed more often than all the other
instruments together. A pilot experiences considerable eyestrain on
a long flight as a result of his continual scanning of the instru
ments. It is for this reason that the instruments should be selected
for readability. Most thought should be given the readability of
the variometer. In order to be read at a glance, the numbers indi
cating the rate of climb should be large with intermediate divisions
not too closely spaced. The needle or other indicating member
should be large enough to be easily visible on the dial or scale. No
precise measure of readability has yet been defined but there are
general considerations for comparing the readability of various
instruments. It is, of course, clear that the readability is improved
by increased sensitivity. A rough criterion of readability is the
total scale length of the instrument for the full range of climb and
sink. Where the scale length is limited, as in the pellet-type in
struments, the readability can be somewhat improved by using a
scale having the graduations for the high rates of climb and sink
compressed. It is also advantageous from the viewpoint of ease of
observation to have a single indicator on a scale. The pellet-type
instruments require the pilot to observe two indicators, the pellets.
In summation, it may be said that the circular scale having a needle
making one turn for the range from 20 to 20 feet per second
most nearly satisfies the various criteria for readability. Further,
a pilot can save himself much eyestrain on a cross-country flight
if he places the variometer on the panel where it is well illuminated
by soft light reflected into the cockpit.
The precision of an instrument is apt to be confused with its
sensitivity. By precision is meant the accuracy with which an instru
ment indicates the quantity being measured. Most variometers,
except the specially compensated diaphragm capillary leak type,
possess inherent errors due to altitude and temperature effects. Since
a variometer is used in soaring merely to detect upcurrents and to
explore them for areas offering a maximum rate of climb, it is rarely
that a pilot needs to know his true rate of climb. Of more impor
tance to the sailplanist are errors introduced into his rate of climb
readings by changes in altitude of his ship. A good variometer
should be statically balanced in order to read correctly over a wide
range of angles of bank and pitch. An error which becomes more
pronounced during steep thermal spiraling is that resulting from
10
Smallest graduation in feet per second.
Capillary Leak
with Liquid
Manometer
o.i
Electrical
0.5
Cobb-Slater
Uncompensated.
is good.
Uncompensated.
is good.
Uncompensated. Affected by
attitude and acceleration.
Precision is poor.
Excellent.
Uniform scale.
Good. Various scales

possible.
Poor even with colored
liquids. Linear scale.
Precision
Precision
Affected by attitude and acceleration. Uncompensated.

Precision is poor.
May be compensated for

altitude and temperature.
Most precise instrument.
Precision
Fair. Scale compressed

for high rates.
Excellent. Scale may

be compressed for high
rates.
5-10
0.5
1.6
Capillary Leak,
Diaphragm Type
Horn
Sensitivity
Type
Readability and
Type of Scale
Lag in
Seconds
COMPARISON OF VARIOMETERS FOR SOARING
Long lag is a definite dis

advantage. Used only where
cost is primary considera
tion.
Requires batteries. Easily

arranged for dual indica
tors. Simple to manufacture.
Best all-around soaring in

strument. Requires careful
adjustment of clearances of
vane and walls. German
manufacture. Patented in
U. S.
Pellets are likely to stick

due to condensation in
tubes. A drying agent (Sil
ica Gel) may be used to
prevent sticking.
Friction in gears and bear

ings necessitates high pres
sure differential, large lag
results.
Remarks
o
^^
cn
OO
INSTRUMENTS
159
centrifugal acceleration in the turn. The pellet and the liquid

manometer variometers are by their design inherently susceptible
to errors of acceleration. During steep turns these instruments may
indicate only one half of the rate of climb they would indicate
in straight flight. When exploring a thermal with either of the
above instruments, one should compensate for their acceleration
errors. Without compensating for the acceleration a pilot might
easily be misled into flying in an area of a thermal which does
not yield the best rate of climb possible from that thermal.
So conflicting are the requirements of a rate of climb indicator
for soaring that it is impossible to find a simple type of instrument
fulfilling all the requirements. If a soaring pilot desires an instru
ment free of altitude and temperature errors as well as one having
a sufficiently small time lag, he is forced by the contradictory nature
of these requirements to carry two variometers, a capillary lead,
diaphragm type and an instantaneous type. For best readability these
two instruments should have similar scales. It has been the practice
of experienced pilots doing contest soaring here and abroad to use
two variometers.
TURN INDICATOR
The turn indicator is shown diagrammatically in Fig. 15A.
A venturi tube through which air flows as the glider advances creates
a vacuum which exhausts the air from the instrument case. The
incoming air impinges upon the buckets of a gyro wheel, rotating
B
A
FIG. 15. TURN AND BANK INDICATOR
it at a high speed. The gyro axis is normally horizontal and athwartship, in a frame which is supported in pivots on an axis which is
horizontal and fore and aft. Turning about the vertical causes the
gyro frame to move about the fore and aft axis against the force

of a restraining spring, thus measuring the rate of turning. The
venturi should, if possible, be so mounted that it may be retracted
when not in use.
To overcome the objections of aerodynamic drag of the large size
venturi necessary to drive a glider turn indicator and the possibility
of ice formations making the instrument inoperative while flying,
an electric type has been developed by Siemens in Germany. In
this instrument the electric motor is incorporated as the gyro. It is
driven by a i2-volt dry cell battery.
A ball bank indicator is usually combined with the turn indi
cator. The face of the combined instrument appears in Fig. 156.
CLINOMETER
A fore and after clinometer (sometimes erroneously called a "pitch
indicator") comprises a triangular-shaped glass tube half full of
liquid, and is graduated to show fore and aft angles. Being affected
by acceleration as well as by angles, it is accurate as a clinometer
only when the glider is flying at a constant speed.
Kollsman instrument Co.
FIG. 16. COMPASS
COMPASS
The compass consists of a pair of mag
nets universally supported on a pivot
within a bowl of liquid. The magnets
are usually carried in or on a float to
reduce the weight on the pivot. The float
carries a graduated card, or rose, which
turns in respect to, and is read against,
a lubber's line. Compensation is provided
for removing the effects of the local
magnetism of the glider.
. i compass is- seen in
Fig.
r- 16.
a
A\ typical
CLOCK
The clock is a useful member of the instrument family requiring
no special description. A standard aircraft clock incorporates a
sweep second hand, and numbers only the hours 12, 3, 6 and 9.
ARTIFICIAL HORIZON AND DIRECTIONAL GYRO
Too heavy for general use on gliders, the Speri y Artificial Horizon
and Directional Gyro could be most useful for instrument flying
in clouds. The horizon is a gyroscopic fore and aft and lateral level
indicator, and the directional gyro is a direction indicator which
is not subject to the unsteadiness of a magnetic compass.
INSTRUMENTS
161
THERMOMETER
A thermometer, although usually thought of as a meteorological
research instrument, is of real value to the glider pilot attempting
cross-country trips. Abrupt changes in temperature indications give
warning of passage from one air mass to another, and even slow
changes in indication may provide useful data.
THE TEMPERATURE GRADIOMETER
This instrument was developed as a device for predicting the
occurrence of convection by means of the temperature field about
a thermal. Unlike the simple thermometer, the temperature gradiometer not only informs the pilot of a temperature discontinuity
in the air mass, but also tells him the direction toward the dis
continuity. The instrument is designed to measure temperature
Photo by Rudy Arnold
FIG. 17. COMPONENTS OF A TEMPERATURE GRADIOMETER

differences across the wing span or fore and aft. It may also be
arranged to measure the total horizontal temperature gradient,
which is made up of the lateral and longitudinal gradients. By
suitable design this instrument can measure easily 0.01 Centigrade
and can make the indication with a time lag of only 0.5 seconds.
Test flights conducted with these instruments over a period of
several years have demonstrated the ability of a temperature gradiometer to predict discontinuities five seconds before they are en
countered, to delineate the zone of lift in ridge soaring and to aid
a pilot in centering his ship on a thermal. As a research tool, the
gradiometer has aided in studying the structure of the invisible
air mass.

A photograph of the components of a temperature gradiometer
is shown in Fig. 17. The streamlined cells are mounted either on
the wing tips or on the nose and tail. The electrical indicator is
mounted on the instrument panel and is connected to the cells by
insulated wires.
HYGROMETER
For meteorological research work, a hygrometer is also carried.
This instrument records relative humidity of the air.
A recommended grouping of the essential instruments is shown
in Fig. 18. An arrangement including the Cobb-Slater variometer
is pictured in Fig. 19.
FIG. 18. INSTRUMENT BOARD GROUPING
FIG. 19. INSTRUMENT BOARD GROUPING

BAROGRAPH
The barograph is a recording altimeter required on soaring flights
for record attempts or contests. In place of the hand of the regular
altimeter, the barograph has a stylus the point of which touches a
rotating, clockwork-driven drum. This drum has a sheet of tin or
aluminum foil wrapped around it and covered with black soot
from a flame of burning camphor. When the point of the stylus is
INSTRUMENTS 163
moved against the drum in the operating position it cuts a line
into the blackening, thus recording the altitude throughout the
flight. Properly sealed by an official, this trace or barogram is the
required definite proof that a sailplane has been aloft every minute
of a cross-country flight. Calibrated before the flight and checked
afterward, it provides an accurate record of the maximum altitude
achieved on a height record attempt. The duration of a flight also
can be measured. A barograph with a drum rotating once every
hour and a range from sea level up to 20,000 feet is most satis
factory.
Photo by Boecker
FIG. 20. THE FEIBER BAROGRAPH
The first barograph developed in the United States specifically
for glider work is the Feiber Barograph, shown in Fig. 20. Its onehour rotation allows careful study of the flight trace. It is of light
weight aluminum construction with large pyralin windows to per
mit reading of the barogram without opening the case. A con
venient locking device allows quick sealing at one point.
Some barographs have been designed for meteorological research.
In addition to recording altitude they have another stylus drawing
a record of temperature and are called "thermo-barographs." A
third stylus is connected with an electric circuit powered by dry
batteries and actuated from a button by the pilot when he wishes
to record the time of something of interest he has observed during
the flight.
CHAPTER VIII
FLIGHT TRAINING
IN INTRODUCING A STUDENT to primary gliding it is well

to show him first the parts of the glider and their functions. At the
same time he should be taught to make the daily "line" inspection.
(See Chapter IV.) By watching the movement of the various con
trols he has an opportunity to obtain an understanding of their
action which will be helpful in mastering their use.
The next step is for the student to seat himself in the cockpit and
fasten the safety belt snugly. The importance of fastening the belt
should be impressed on him so that he will form early the habit of
doing so as his first action on seating himself in any aircraft. It will
give him a feeling of security, a "oneness" with the glider, and may
prevent serious injury in case of accident.
Placing his feet on the rudder bar or pedals, as the case may be,
the student is taught how to steer. With right foot forward he can
look back and see the rudder moved to that side. It is well to point
out that its action in the air is the same as that of a boat rudder in
the water. If his winter sports have included much sledding, he
must learn that this steering is exactly opposite. To turn to the
right: right foot forward. To turn to the left: left foot forward.
This soon seems very natural.
Equally natural is the use of the stick for lateral and longitudinal
control. When you look to the right you are likely to incline your
head that way. So you move your stick to the right to bank to the
right. When you look up or down you pull your head back or
bend it forward which corresponds to the action of the control
stick in climbing or diving. Another way to place its action in
your mind is to try to think of it as being rigidly fastened in its
socket. When the left wing drops you move the stick to the right
to bring it back up as if you had twisted the whole glider around
that way.
As the student moves the controls to learn their use, he can begin
another valuable habit, and that is, after fastening the safety belt,
always to move the controls as far as they will go in all directions
before taking off. This will show whether they are unobstructed
164
FLIGHT TRAINING 165

and moving freely, which of course is essential to safe operation.
Later on he will also check the operation of the brake and the re
lease, but it is better not to confuse his mind with too many details
at first.
Before making any tows, it is advisable to let the student learn
the use of the stick by sitting in the cockpit with safety belt fas
tened and balancing the stationary glider in a steady breeze of 15
to 20 m.p.h. To prevent the glider's blowing over it is advisable to
tie the nose to a stake with 5 or 6 feet of rope if the wind is strong
or gusty. If the glider is balanced on its wheel it is possible even to
feel some response to the rudder as it twists the ship slightly from
side to side in a strong wind. A glider should never be left faced
into a strong wind unless it is well staked down or someone is
seated in the cockpit. There should be no obstructions, not even
persons standing in front of the glider, as this disturbs the airflow
and interferes with the action of the control. The student here can
be taught to look straight ahead at the horizon and watch it for
keeping his wings level instead of looking sideways at the wing
tips. One or two i5-minute periods of this practice are usually suffi
cient as a preliminary to ground tows.
Although other methods of towing have been used with success
for primary training in single-seater gliders, the automobile tow
method is by far the safest and most practical where large, level
fields are available, so it alone will be described in detail for the
first instructions. With it an experienced instructor can have real
control over his student and prevent him from getting into trouble.
The equipment needed includes a sturdy car with good power
and pick-up and a i5o-foot rope. The car should be a roadster or
other open type. Full visibility is very important; a closed car with
fixed top should never be used. On the rear end of the car there
should be an approved type of release. Fastened to this tripping
mechanism should be a cord or light piece of rope which at all
times should be close beside or in the instructor's hand.
First tows across a level field or airport should be made in a calm
or into a wind of not over 5 m.p.h., such conditions usually being
found in the early morning or late afternoon. In primary instruc
tion in a single-seater glider it is advisable to have the stick secured
forward so that it cannot be pulled back past neutral. This check
is best made of stranded control cable with a turnbuckle for ad
justment. At first, it is best to check the elevators so that a full
backward pull on the stick will bring them only to a point just
below neutral. The turnbuckle is then made safe with wire to
assure this adjustment remaining constant.

The purpose of this check is to prevent the glider's lifting into
the air either from a sudden gust of wind or from too much towing
speed during the first tows across the airport. It also gives the
student confidence in knowing that he cannot get into trouble if
he makes a mistake. The stick should then be held as far back as it
can gonot pulled hard back, but held neutral in this position.
Before making the first tow, the student makes sure that his
safety belt is fastened, his feet are squarely on the rudder bar and
his right hand is holding the stick centered. The instructor makes
sure the rings at the two ends of the i5o-foot rope are properly
fastened in the release mechanism on the glider and on the tow
car. He keeps the car in low or second gear and his hand on the
release rope.
If a third person is available, it is helpful to have him run along
and hold the wing level until the glider has sufficient air speed
for lateral control. If not, the student must start off with his stick
held all the way to the side of the high wing. He is told to look
straight ahead only and to concentrate on keeping the wings level
and the glider straight behind the car. His grip on the top of the
stick should be firm but relaxed.
The usual fault of most beginners is to overcontrol the rudder.
When the glider swerves to one side of the car, the student should
push the rudder in the opposite direction but should be ready to
bring it back to neutral as soon as the glider is once again in line
behind the car. Failure to do this in time will cause the glider to
swing to the opposite side. The usual result of attempting to correct
this by overcontrolling the rudder will result in zigzagging across
the field. This in turn will exert tipping forces on the glider
which will make it difficult to keep the wings level. Remember,
when the glider has swung too far to the left, push right rudder
until it is once again in line, then push enough left rudder to keep
it there. There is always this double actionmovement of the con
trol to change direction and reverse movement to hold it in the
desired course.
The same applies to learning aileron control for balancing the
wings. As the student looks straight ahead and feels himself lean
to the left as the left wing drops, he must push the stick to the
right, all the way if necessary, until he feels the wing come up. As
soon as he feels himself once again sitting level he must move the
stick back to the left to hold that position and prevent the right
wing dropping. It is very helpful for the student to keep checking
himself to make sure that his right hand holding the stick is re
laxed and that his grip is a light one allowing him to feel what is
FLIGHT TRAINING 167

going on. You do not need any appreciable degree of physical force
to fly, but you must have a sensitive feel and good co-ordination.
This co-ordination of eyes, sense of balance, and muscular con
trol soon comes with practice. Despite repeated verbal directions
from the instructor, the student's grasp of flying is largely self-taught
and therefore never leaves him. Learning directional control with
the rudder and lateral control with the ailerons may require many
tows up and down the field with frequent stops when the zigzags or
wing hangings become too severe. The instructor should be patient
and must be ready at all times to trip the release which may pre
vent damage to a wing tip when the student has swerved sharply
one way and the resultant slack in the towrope suddenly has been
taken up causing the glider to swing quickly in the opposite direc
tion. Learning to fly this way can find a parallel in learning to
swim. Despite continuous effort to stay afloat you keep sinking
until suddenly you find that you can stay up with no effort and
don't know quite how you learned to do it. Similarly the gliding
student finds that he can keep the glider straight and level with
very little effort.
When the instructor has satisfied himself that the student really
has caught on to these two controls by making several tows he can
increase the speed of the car until the wings of the glider are just
lifting its weight. With the average primary or utility glider this
is at an air speed of about 23 m.p.h., so if, for instance, the towing
is being done into a wind of 8 m.p.h., the car should be driven at
15-20 m.p.h. A slight readjustment of the cable check on the con
trol stick may be necessary to allow the glider to skim off the
ground a few inches.
With this first leaving of the ground, the student begins to ex
perience the great thrill of motorless flying. The bumping across
the field becomes less and less as the wings gradually take up the
load and the craft becomes airborne. With this increase in speed
there is also increase of control so the student will find it less effort
to keep straight and level. The importance of keeping the wings
level when in flight close to the ground should be emphasized.
As the student becomes more at home in controlling the glider,
the instructor can loose the check still more and allow flights up to
2 or 3 feet. With the check still on there is possibility of danger
in climbing too high as the student might push the stick too far
forward causing the glider to dive and then not have sufficient
control to level out before hitting the ground. For this reason it
is advisable to undo the check entirely as soon as the student has
shown sufficient aptitude. If the student should pull back too far

or be lifted suddenly to 10 feet or more by a gust and continues
to hold the stick back while the check is still on, the instructor can
allow the glider to settle to the ground by gradually decreasing the
speed of the tow car.
As he skims a few feet off the ground, the student can turn most
of his attention to mastering the fore and aft movement of the
stick as he now keeps straight and level almost instinctively. The
best way to attain an equal proficiency in elevator control is to pull
back gently when the glider has sufficient flying speed, climb to a
height of 5 feet and then concentrate on trying to hold that ele
vation without touching the ground or climbing higher while
being towed the length of a long field. At the start the student
must be cautioned never to take off until he is sure that his wings
are level and his course is straight behind the car.
It is well for the instructor to give a hand signal when he is about
to slow down at the end of the field so that the student can nose
down slightly and land. The correct procedure in landing is to pull
back gently just a little on the stick when a few inches off the
ground so that the wheel or central part of the wooden skid touches
first. In landing, the student must be taught to look from 100 to
300 feet ahead so as to be able accurately to judge his height above
the ground. If he looks down directly in front of him he sees the
ground rushing by and cannot form an accurate judgment of ele
vation.
After 8 to 10 tows where consistently level heights of around
5 feet are maintained and good landings made, the average student
is ready to go higher and make his first free flight. The best way
to instruct him in this is to tell him to climb to about 15-20 feet
and prepare to release when the instructor signals him down with
his hand. He should be warned not to hurry but to nose down a
bit and then pull the release ring. Before this flight and on all
subsequent flights, he should keep his left hand on the release at
all times while being towed. After releasing he should glide down,
keeping steady flying speed until the slight leveling out to land.
This speed he can judge best by the amount of wind on his face
and the "solid" feel of the controls. For primary instruction this is
preferable to the use of an airspeed indicator which may become
inaccurate from being clogged by grass seeds, dirt or moisture
while towing over the ground. It is better at first to develop a sense
of flying "feel" rather than a dependence on instruments.
After his first free glide, which constitutes his second real mile
stone and thrill of his gliding career, the student can be allowed
to make higher and higher tows until he has reached the maximum
FLIGHT TRAINING 169

possible with the ir,o-foot rope. This is about 120 feet and seems
almost directly over the tow car. From this height the first shallow
turns can be taught. This is the most critical stage in gliding in
struction where the advantages of two-seater training become most
apparent. Lacking this equipment, instructions with a single seater
should be given with great care.
The student should be told about the causes and dangers of
stalling as a result of insufficient flying speed and how easy it is to
stall and fall off in a turn which requires slightly more flying speed
than straight flying. The first turns should be made from a height
of about 100 feet and should be only i5-20. If there is any wind,
it is advisable to tow slightly crosswind so that the landings are
made directly into the wind. The student should be cautioned not
to pull the stick back causing the glider to stall nor to push it
forward causing it to dive as he moves it to one side to bank. In
the average, well designed glider only a slight amount of rudder is
needed in the direction of the turn. It is wise to have him make
each maneuver separately: get sufficient flying speed on the ground,
take off, climb, nose down slightly, release, steady flight at proper
speed for a moment, turn, level out, land. In turning one should
always lead off with the stick to bank, followed immediately by
the rudder. Too much bank for a given radius of turn will result
in a slip. Too much rudder will result in a sideways skid and pos
sible stall. All turns should be completed at least 20 feet above the
ground so that there never will be the danger of a wing tip touch
ing the ground while the glider is in flight.
From the first gradual turns right and left he progresses to 90
and finally 180 turns with the i5o-foot rope. Greater towline
length may be preferable for two-seater instruction or on excep
tionally large fields. Although first turns should be made in a calm
or with a slight breeze, it is advisable to have well advanced stu
dents become accustomed to turning in winds of up to 20 m.p.h.
Downwind landings should not be attempted in such strong winds,
however. In making 90 and 180 turns the instructor should take
care to give the student as much room as possible with regard to
obstructions bordering the flying field. At first he should signal
when to release, but later he should tell the student to use his own
judgment. If the glider is equipped with a brake the student should
keep his hand on it when landing. If he is overshooting and in
danger of colliding with a fence or other obstruction he must pull
on the brake and push the stick hard forward. Unless he is landing
on wet grass, this will bring the glider to a quick stop.
After the student can make consistently good 180 turns in

either direction, he is ready for higher tows with a longer rope.
The student should be cautioned on all high tows not to pull up
steeply until over 200 feet high so as to have sufficient altitude to
recover in the event of the rope breaking. With the 5oo-foot length
the student can climb to approximately 400 feet which is ample
to make a 360 turn and landing in the center of the field. The
best way to instruct him to do this is to tell him to make a 180
turn after releasing, fly downward over one side of the field to the
other end and then make another 180 turn in the same direction
and come in for a landing in the center of the field.
3ooft. ki
A 360 TURN TO THE LFFT

He must be cautioned to use his judgment in deciding when to
turn so that he will not overshoot or undershoot the field. There is
danger of the former during good thermal conditions which may
prolong the flight. It is well under such conditions for the instructor
to make a 360 flight first both to determine the advisability of let
ting the student try it at that time and to show him how he should
do it. If he should undershoot slightly and run into danger of not
quite getting over the fence or other obstruction he should be told
to dive toward the bottom of it and then pull over it rather than
stretching his glide which may cause him to stall down on top of it.
After he has made several of these flights in either direction he
can try a figure eight, making the first 180 turn in one direction
and the second in the opposite direction. Soon the handling of the
glider in the air will become quite natural and the student can use
FLIGHT TRAINING
171
his own judgment about when to turn and can concentrate on

making precision landings. With a little practice he should be able
to land and stop without the use of the brake within 100 and later
within 50 feet of the airport circle or other designated mark.
Between lessons, especially if they are spaced far apart, a student
should be required to "check out" by going through the preliminary
stages of ground skimming and straight flights from low altitude to
demonstrate his ability to progress further that day.
At this stage it is well to demonstrate a forward slip and slight
fishtailing, the former for losing height without increase of speed
and the latter for reducing excess speed close to the ground. A slip
is a valuable maneuver for losing height at a steep angle after mis
judging or coming in too high purposely to be sure to clear obstruc
tions. By banking to the left and applying right rudder, the fuselage
is skidded sideways exerting considerable drag without impairing
the lift of the wings. Naturally this maneuver is useless with a
primary glider which has no side keel area to make the necessary
resistance. Fishtailing is kicking the rudder hard from side to side
while keeping the wings level. Again, these maneuvers best can be
taught first in a two-place, dual-control glider.
It will be of great assistance throughout the course of training
to have the instructor demonstrate all the various maneuvers. It is
also helpful to have a class of several students with some of those
who are not flying riding in the tow car and others watching from
the sidelines trying to analyze the mistakes of the student in the
glider.
Where fields are too small or sloping to permit automobile tow
ing, successful training has been done by the shock-cord or "bungee"
method. An outstanding club where initial training is done entirely
by this means followed by winch towing is the London Gliding
Club. Short "slides" followed by skimming flights are given to the
students after the same introduction of balancing the stationary
glider in a wind that precedes automobile towing.
Using the general procedure of shock-cord towing outlined in
Chapter V, the power of the first tows is greatly reduced by less
man power and shorter stretch of the rubber rope so that upon
release the glider at first does not leave the ground but has sufficient
speed to permit lateral control for a short distance. With this system
the student is instructed at first to keep his rudder centered with
the foot bar straight across for the first few skimming flights.
It is impossible to give exact details for this type of towing due
to the slope and surface friction of the ground, the life and conse
quent elasticity of the shock cord, the pull of the towing crew, and

the force of the wind, which are all variable factors. The instructor
should use the general information about automobile tow and
shock-cord launchings and proceed cautiously so that an inexpe
rienced student will not be catapulted into the air until he has
had sufficient ground slides and skimming flights.
PRIMARY SOARING INSTRUCTION
For the most part, glider pilots have made their first sustained
flights and won their "C" licenses for 5-minute flights above point
of release over a slope soaring site. This is the easiest type of soaring,
still most commonly done. If possible it should be preceded by
instruction in a two-seater glider or in a light airplane teaching
stalls and spins and recovery therefrom.
For slope soaring one usually must have a steep ridge of 45 or
more and several hundred feet high facing the wind with safe
landing areas on top or bottom, preferably both. If taking off on
top of the ridge from a field 800 or more feet long winch towing
is preferable. If the field is too small for winch or auto tow either
hand or auto shock cord must be used. This latter launching
should never be used for a student making his first soaring flight
unless he has had plenty of practice with it on a large, level field.
He should also have had winch towing practice if a winch is used.
A utility type glider equipped with airspeed indicator, variometer
and altimeter is proper for preliminary soaring. A wind of 15-20
m.p.h. is usually preferable for first slope soaring.
After releasing from the launching he should fly straight out until
he is in the area of rising air in front of the slope. This he usually
will feel, but also can notice by the variometer needle moving up,
indicating that the glider is climbing. He must be cautioned not to
pull up the nose of the glider when encountering the lift as this
may cause a stall and in any case increase the glider's sinking speed.
Instead he should fly at a steady speed a few miles per hour faster
than normal gliding in quiet air over an airport. This will mean
about 30 m.p.h. for the average utility glider. The added speed is
for better control, as some turbulence may be encountered.
As soon as his glider is in the area of rising air the student should
turn in the direction that will allow him the longest straight flight
before reaching the end of the ridge. When he does come near this
point he should make a 180 turn into the wind away from the
ridge and continue parallel to the ridge down to the other end
where he will turn in ihe opposite direction, again into the wind
and ,-iway from the ridge. While flying a course parallel to the ridge
the glider will be pointed sideways into the wind, that is, it will
FLIGHT TRAINING 173

be "crabbing" sideways to hold this course while flying crosswind.
As a result its ground speed is often very slow. It is possible that he
soon will rise to a height of 500 to 1000 feet above the ridge where
the air will be smoother and the area of lift wider. After half an
hour he should fly out over the valley away from the area of lift
and glide down to a landing in a designated field. If he stays up
longer than this he is apt to become overtired which will affect his
SLOPE SOARING
judgment adversely. It is advisable before a student's first soaring
flight to have the instructor make a flight first showing just how
it should be done.
Although in the past slope soaring has nearly always preceded
thermal soaring, the latter, which offers far greater possibilities for
real performances, is being used more and more to obtain primary
soaring experience. The reason for this is that many gliding clubs
now exist in level country far from any mountains. Before trying
this type of soaring it is even more important, if possible, to have
dual, two-seater glider or lightplane instruction. It is necessary to
learn, in addition to stalls and spins, how to make a continuous
turn or spiral of comparatively narrow diameter. This can well be

taught at a reasonably high altitude of several thousand feet by
gliding down in a two-seater glider after releasing from airplane tow
or gliding down in a lightplane with the engine throttled. The
instructor should teach a student how to make turns where the
glider is banked more than 45 and the elevator and rudder con
trols become reversed.
For the first thermal soaring over level country it is advisable to
have a large field, allowing winch or auto towing for at least 4000
feet. This will make possible heights of 800 feet or more. Most
thermal "bubbles" are comparatively narrow and weak at low alti
tudes and can rarely be used to climb when encountered below
500 feet, especially by an inexperienced pilot. Convection, or
thermal activity, is best in the summer months and rarely is strong
enough for soaring before ten o'clock in the morning.
As in other phases of training it is best for the instructor to make
a flight to show the student how it should be done. If he is familiar
with the local conditions he probably will know where thermals
usually start. It may take the student a number of tows before he
encounters one of these invisible bubbles of rising air. When he
does, usually he will feel a slight turbulence. However, this is not
always so and in any case he should pay close attention to his
variometer and air speed.
Sometimes he may glide into a weak thermal which will bring
the variometer reading up from the normal glide of 3 feet per
second to the zero mark. This means that the thermal is rising at
3 feet per second, so by spiraling to stay within it the glider just
maintains altitude. One of the author's students encountered such
a condition after releasing at 700 feet and by careful spiraling man
aged to stay up 15 minutes on his "C" flight although he rose only
slightly above this height. Others have climbed several thousand
feet on their first thermals. While spiraling in thermals it is well
to keep an eye on the position of the glider with respect to the
landing field as the thermal acts very much like a balloon and will
drift with the wind. It may be necessary to leave it to be sure to be
able to glide back to the field. Wind is not always present on days
of good thermal activity.
CHAPTER IX
SOARING TECHNIQUE
OOARING FLIGHT CAN BE DIVIDED into three principal

categories: endurance, distance and altitude. Of the three, endurance
soaring is the least important.
It is the aim of every experienced soaring pilot continually to
improve his knowledge and skill until he is able to make record
performances. A new record with its attendant publicity is always
a help to the progress of motorless flight and this is the chief value
of endurance soaring, which otherwise is only a "glorified form of
flagpole sitting" proving nothing more than the constancy of the
wind and the physical endurance of the pilot.
Without exception all endurance soaring records have been made
over slope soaring sites. Most of them have been made over sand
dunes on steady sea breezes. From the time a pilot makes his first
half-hour soaring flight he soon progresses to flights of 2 and 3 hours.
His first endurance goal is the 5-hour requirement for the "Silver C"
license. Beyond that he has to look for incentive to set new state,
national and international records. Many states have yet to have a
record established but both the United States and the world records
are now so high that any attempt to better them will involve night
flying. These records are divided into two categories for singleand multi-seater gliders.
Before a pilot attempts an endurance flight of even so much as
12 hours he must be in excellent physical condition and should
make flights of 7 or 10 hours which will teach him many things
that are essential to the success of later record attempts. First of
these is that he must be adequately clothed. Sufficient warmth is
even more important than a comfortable, well-padded seat. He
should carry a supply of food and drink. Experience on several long
flights has taught us the value of eating something such as a choco
late bar or cookies every hour to keep up energy. Oranges are
valuable as thirst quenchers. If a flight is continued on into or
through the night hours it is wise to have a thermos of hot coffee
or other stimulant to help keep awake. Caffein pills may also help,
but these should be used with care. Of great importance are sanitary
'75

considerations. On any flight of more than an hour's duration,
especially in cold weather, the pilot should carry an empty bottle
with wide neck and with watertight screw cap. In a flight of 5 hours
or longer this is an absolute necessity, because a full bladder can
cause extreme physical discomfort affecting flying skill and judg
ment. There is the very real danger of a possible hard landing, or
crack-up, when a sudden blow easily can cause a full bladder to
burst with fatal results. Although it is not generally known, many
deaths have resulted because of this condition from automobile
and airplane accidents which otherwise would not have caused
serious injury.
Any night soaring should be preceded by a number of hours of
instruction in night and instrument flying in airplanes. Unless a
pilot has this experience he may get into serious trouble as he
probably will find that he cannot adjust his judgment on his first
night flight, unless, of course, there is bright moonlight. Even if
this is so, thick clouds may cut off all light so that he will have to
rely entirely on his instruments to maintain level flight.
These must include an airspeed indicator, variometer, altimeter,
and turn and bank indicator. A clinometer, compass and clock are
also useful. All instruments should be illuminated by radiolite
figures and pointers or by indirect lighting, preferably controlled
by a rheostat. If available, a 5-meter transceiver radio set can be
of real value on such a flight. To comply with government regula
tions, navigation lights must be carried: a green light on the right
wing tip, a red light on the left wing tip, and a white light on
the tail.
For any night endurance soaring attempt there should be a
number of guiding lights spaced as evenly as possible outlining
the top and bottom of the ridge. In addition, the landing field or
area should be outlined with small lights such as lanterns. If am
floodlights are available, they should be used with care as they are
apt to blind the pilot. As for all other soaring record attempts, the
glider must carry a barograph installed and sealed by the qualified
official who witnesses the take-off and landing.
Certainly one of the most important considerations is the choice
of a site for the record attempt. Some of the best places in America
are the sand dunes in the vicinity of Frankfort, Michigan, which
offer almost unlimited facilities for landing on the beaches. Even
on the darkest night it usually is possible to see the glimmer mark
ing the sin 1 and to land alongside it.
The question of air traffic with two or more gliders slope soaring
over the same ridge is a serious one, especially if the ridge is limited
SOARING TECHNIQUE 177

in extent. Before a number of pilots take off for soaring on the
same ridge they should all agree to definite traffic rules to lessen
the ever-present danger of collision. In passing a glider coming in
the opposite direction it usually is best to keep to the right. In
overtaking another glider flying at the same altitude it is best to
stay on the side closest to the ridge where the lift is strongest. If
flying at approximately the same altitude as several other ships, it
is most important to follow the rule of always turning away from
the ridge. There have been head-on collisions between two gliders
where one pilot turned suddenly in the opposite direction to that
expected by the pilot following close behind him.
Endurance soaring for new records had best be done in the twoseater category, as demonstrated by the last two international rec
ords. It is doubtful if the present mark, a flight of more than two
days and nights, could have been made solo. With two pilots reliev
ing each other in shifts, the chances of making a new record are
greatly increased without the danger of the pilot falling asleep.
The greatest value to be derived from endurance soaring over
ridges, aside from record setting, is the building up of general flying
experience. Hours spent in the air piloting under any sort of condi
tion can be of value in building up the combination of skill and
confidence needed by a good all-around pilot. While making endur
ance flights a pilot can best prepare himself for the more important
types of soaring flight by perfecting his turns and generally concen
trating on making all his flying smooth and effortless.
DISTANCE SOARING
Although it is possible to use a secondary or utility glider for
establishing endurance records by slope soaring, an intermediate or,
preferably, a high performance type is needed for long distance or
high altitude soaring on thermal upcurrents. Before attempting a
cross-country flight with one of these sailplanes a pilot should first
make a number of flights to familiarize himself with his ship and
feel at home in it. In getting used to it he should practice enough
landings to a mark in varying atmospheric conditions so that he
can land safely in a small field if need be.
The easiest way first to practice the thermal soaring technique
essential for either distance or altitude soaring is from airplane tow.
By this method the instructor piloting the airplane can tow his
student in the sailplane to an altitude of 3000 feet or more where
the thermal upcurrents are wider and stronger than they are closer
to the ground. With a sensitive variometer installed in the airplane
the instructor can tell when he is in rising air and signal to the

student just when to release. Probably even before releasing the
student will notice his variometer needle registering the fact that
the sailplane is rising. To stay within the invisible boundaries of the
rising bubble or column of warmer air he must immediately bank
his sailplane and begin a steady turn or spiral.
The ability to make smooth, continuous spirals is the most impor
tant single accomplishment in advanced soaring flight. In making
shallow turns of consequently rather large diameter the pilot keeps
his nose far enough down to fly 4 or 5 m.p.h. above stalling speed,
leads off with his stick banking the sailplane and immediately folCORRECT
STRAIGHT - LEVEL
SHALLOW LEFT TURN
5TEEP LEFT TURN
INCORRECT
STRAIGHT-LEFT WING LOW
LEFT TURN-SUPPING
LIFT TURN-SKIDDING
VARIOUS FLIGHT ATTITUDES As SHOWN BY THE TURN AND

BANK INDICATOR
lows with enough rudder to hold the ship in the turn. He then holds
his controls so that he continues to turn, being careful to coordinate
his stick and rudder so that the ball of the ball bank will remain
centered at all times. If he has too much bank or too little rudder
for a given left turn the ball will roll to the left indicating that
he is slipping down out of the turn. To correct this he should move
the stick to the right to reduce the bank or push his left foot far
ther forward to apply more rudder until the ball is again centered.
As he becomes more experienced he usually will use a combination
of the two controls to smooth out his turns. As thermal upcurrents
are often quile turbulent he must be constantly on the alert to
keep flying as smoothly as possible.
In the event that thermal soaring is being done without a ball

bank in an open cockpit sailplane the pilot must rely on his sense
of feel while watching the variometer. If he has loo little bank or
too much rudder for a given left turn he will feel a blast of air on
his right cheek indicating he is skidding out of the turn. If equipped
with a ball bank this will be shown by the ball rolling to the right.
Correction is made by steepening the bank or reducing the rudder
until the ball is centered again. A good rule to remember for open
cockpit flying without the ball bank is to correct by pushing rudder
on the side you feel the blast of excess slipstream indicating that
you are skidding or slipping.
After becoming proficient in making smooth, shallow spirals the
student should practice the steep turns often necessary for climbing
in thermals of narrow diameter or staying in the center of large
thermals where the lift is strongest. As this requires more skill and
can more easily get the student into trouble, it should first be prac
ticed at a reasonably high altitude of over 1000 feet. If airplane tow
or high winch launchings are not available for practice flights it is
wise to get some steep turn instructions in a light airplane. Dual
instruction in a two-seater sailplane released from airplane tow is
the ideal method.
Before making steep spiral turns it is well first to have an under
standing of the forces involved. In speaking of steep turns we con
sider those where the wings of the sailplane are banked at an angle
of more than 45 with respect to the horizon.
In making a turn a glider produces a centrifugal force which
tends to force it out of the turn. The shorter the radius of turn,
the stronger this outward pull becomes. For a correct turn the
angle of bank must be so that the component or the result of the
combined pull of gravity downward and centrifugal force outward
exerts a pressure vertically to the wings, holding the glider in the
turn. The pilot must use the correct amount of bank for a given
radius and air speed to make a proper steep turn or continuous
spiral.
In steep turns the action of the controls becomes such that back
pressure on the stick must be applied to hold the ship in the turn.
In starting a steep spiral it is best to nose down to a speed 5 to
10 m.p.h. in excess of normal cruising speed and then roll into the
turn gradually by proper coordination of stick and rudder. As the
ship banks beyond approximately 30 degrees the stick is pulled
back sufficiently to keep the ball centered, very little rudder being
used. Constant adjustment of stick and rudder may be necessary
when spiraling in turbulent air.
In learning to master tight spiraling it is easy for the student to

stall the sailplane and fall out of control. If too much rudder is
used at the beginning of the turn a skid will result which may cause
a stall. If he is on his toes and is sensitive to the feeling of sloppy
controls when a ship stalls he will be able to recover before a spin
develops by straightening out and nosing down until flying speed
is regained.
A common mistake of students making tight spirals is to pull
back too hard on the stick and either forget about the rudder or
apply rudder on the down side. The usual result is a sudden stall
followed by a fast spin as the nose drops. It may sometimes take
as much as 200 feet to recover, so it is obvious that a beginner
should never make steep turns at low altitudes.
Another common cause of stalling and falling off is pulling the
stick back too soon at the beginning of a turn. This will cause the
air speed to drop close to the stalling point before the turn is well
started and the additional drag on the turn will bring about the
complete stall.
A complete understanding of stalls and recovery therefrom is
essential before any attempt at advanced soaring. It is a common
thing to have a sailplane stall a number of times during a distance
flight due to flying slowly in turbulent air. A gradual stall in
straight-away flight in the average, well designed sailplane first
makes itself known to the pilot by the sloppy feeling of the con
trols. The stick can be moved some way before there is any feeling
of response, which is then quite sluggish. If the pilot is experienced
he will immediately push the stick forward and regain full flying
speed.
If the stall is abrupt from overcontrolling, a badly executed turn,
or turbulent air, the sailplane may fall off into a spin in which the
nose is down at a steep angle and the whole ship rotating as it falls
with the tail making a turn of larger radius than the nose. A well
designed ship will come out of a spin after not more than two
turns if the pilot takes hands and feet off the controls. It will then
fall quickly into a nose dive in which the flying speed and control
are regained. Usually it is possible to recover from a spin faster
than this by centering the stick and kicking rudder in the direction
opposite to the spin which will quickly stop the rotation. As flying
speed is regained in the resulting dive the pilot pulls the stick gently
back to raise the nose to the normal gliding attitude.
A type of stall that is apt to be disconcerting to a beginner some
times occurs in thermal soaring where, due to the large wing span
of the sailplane, one wing may sweep out of a thermal into a down
draft while the other remains in the rising air. This may cause the

high wing out of the thermal to stall, resulting in the sailplane
tailing off in that direction despite an indicated air speed well in
excess of a stall.
In preparing for a cross-country soaring flight a pilot must equip
himself just as for an endurance flight, with a few additions. He
should have a parachute, preferably a back pack for convenience
and comfort. His seat should be equipped with an air cushion or
other suitable padding and he should have sufficient clothing. On
a day when it may be too hot for more than shirt sleeves on the
ground it will seem quite cold at 5000 feet. A white cap with sun
visor will prevent danger of sunstroke. Other considerations as to
food and comfort are the same as outlined for endurance soaring,
as a distance flight may easily last more than 5 hours. His instru
ments should include one or two variometers, an airspeed indicator,
ball bank, altimeter, compass and barograph.
A soaring pilot's first cross-country flight may be an attempt to
qualify for the 32-mile distance requirement of the "Silver C"
license. Before making any flight of this or greater distance, he
should study carefully the features of the terrain to be flown over
so as to know what to expect both to avoid dangerous areas and to
be able to navigate properly. Aerial navigation is a subject the full
explanation of which has taken entire books the size of this one,
but the navigational problems of the sailplane pilot are not very
complex. However, although they are comparatively easy to explain,
they may take some time for the student to master unless he has
an expectionally fine sense of direction and orientation.
First of all he must have a good aerial map showing such features
as towns, mountains, rivers, lakes, railroads, and principal highways.
The best maps for this purpose are the Sectional Airway Maps
printed by the Civil Aeronautics Administration in Washington,
D. C. These have a scale of 8 miles to the inch and 1000 foot con
tour changes indicated by different shadings of color. All unnecessary
and confusing details have been omitted and the important essen
tials made very clear. Lacking one or more of these maps, the best
of road maps, such as those distributed by the Esso Touring Service,
may be quite adequate, especially if the pilot has no definite des
tination and wishes simply to fly straight down wind as far as pos
sible. The chief disadvantage of these maps is that they do not
include the railroads.
Before taking off the pilot should check his wind direction and
draw a line on his map indicating his probable course straight
down wind. It is helpful to mark lo-mile intervals along this line
so that he will be able to check his progress cross country. Also

marked along the line should be the compass heading he wishes
to fly. This can be found easily by laying a transparent protractor
calibrated from zero to 360 over the map. The center point should
lie on the course line where it intersects a line of latitude or longi
tude. If one of the former is used the East-West or 9o-27o line
of the protractor should exactly coincide with the horizontal line
on the map indicating the latitude. The magnetic course can then
be read off directly in degrees. This must then be corrected for
deviation and variation.
Deviation is a magnetic error caused by attraction of metal parts
in the sailplane. As most sailplanes are largely of wood construction
this rarely needs to be considered. However, it is well to have the
compass compensated and errors, if any, listed on a small card.
Variation is the difference between true North and the magnetic
pole. It is indicated on the Airways Maps by dotted red lines listed
at the edge as a certain number of degrees plus or minus. Variation
West is plus and East is minus. This can be remembered by "West
is best and East is least."
If the wind on the day of the cross-country flight is from the
northwest, the down wind course will be read off the map at 135.
If there is a deviation error of 2 this is subtracted to make 133.
If the variation in the particular part of the country is 9 West
or -|~9", the course to be flown is 142.
Whether a road map is used to navigate in the air or not it is
advisable to have one of the area in which the landing will be
made and leave a duplicate with your ground crew. This will
greatly facilitate letting them know your exact location when tele
phoning after the landing. It is helpful, too, to write down the
number of the telephone at the starting site on the road map car
ried in the sailplane.
It is best to start from a well tested site where thermal activity is
known to be excellent on days of good convection. If this is a
place such as the Warren Eaton Site at Elmira, the flight may start
with local slope soaring as soon as the wind is of sufficient velocity
and blowing up the ridge. This may be as early as eight o'clock,
but the thermals are seldom strong enough here to start cross
country before nine-thirty or ten o'clock.
When a thermal upcurrent is encountered while slope soaring,
its presence is usually indicated by a slight turbulence and a definite
rise in the rate of climb as shown by the variometer. Although
there may be one or two good cumulus clouds, it is well to wait
until more form, indicating steady convection at regular intervals.
The start cross country should not be made until the pilot has

climbed to the cloud base, which may be from 3000 to 5000 feet.
In no case should he start from a ridge site such as Elmira unless
he is at least 3000 feet above his take-off. Otherwise the downwash
effect on the lee side of the mountains may cause him to lose so
much altitude that he is forced to land after covering only a mile
or two.
It is well to consider carefully the adverse effects of downwinds
on the lee side of mountain ridges, as they sometimes are very dan
gerous for any type of aircraft. The author can remember two cases
LAR.ce
CUMULUS CLOUD
(60*4)
(Avoid)
WOOOS
SHAM?
WOODS
OPM
f=l*i.DS
CROSS-COUNTRY SOARING
where they almost caused him to crash. One was in Virginia while
slope soaring about 500 feet above a 2ooo-foot ridge The altitude
was 4500 feet above sea level and cumulus clouds were coming by
about 1000 feet over the slope. By watching one of these too care
fully while endeavoring to catch the thermal forming it, he failed
to see that the sailplane was drifting to the lee of a spur of the
irregular ridge. Suddenly the variometer registered a very rapid
descent of more than 10 feet per second. Only by diving at 65
m.p.h. was he able to clear the ridge by a scant 20 feet and reach
the safety of the windward side. On another occasion on a long
mountain range in Pennsylvania he was forced to fly to the lee of
the ridge as the wind changed sufficiently to render slope soaring
precarious over a long belt of unbroken forest. On the downwind
side there was a possible landing field. As the sailplane flew toward
it at a fairly high speed of 50 m.p.h. to give adequate control, it

suddenly was caught in the downdraft so violently that both wing
tips flexed sharply and the cockpit cover was torn loose, the wood
splitting and the screws pulling out. If the sailplane had not been
well designed with adequate reserve strength, it would probably
have broken up in the air by this terrific jolt. It is best always to
keep in mind the wind direction and to remember that with suffi
cient velocity there will always be downdrafts in the lee of ridges
and that these are likely to be turbulent and dangerous.
After the first spiral climb to the cloud base the thermal should
be left before entering the cloud, especially if the pilot is not expe
rienced in cloud flying and equipped with the necessary instru
ments. In starting off downwind it is best to increase speed to get
through the adjoining downcurrents quickly before much height
is lost. After that, in air that is neither rising nor falling it is best
to fly at minimum sinking speed to conserve precious altitude. It
Pis-fane.*. Soaring
^'''gA/-
- >/e ;/&> ,/*-Y., r0
P'K*tor,> At -/$&
Afrit 3, /<?3S" - e;n &.&*r
BAROGRAM OF DISTANCE FLIGHT

is often advisable at first to head off at an angle instead of directly
downwind where the downcurrents are apt to be strongest. Then
a downwind course should be held toward the nearest cumulus
cloud that seems to be growing, indicating strong lift beneath it.
Of great importance for any successful cross-country soaring
flight is an ability to differentiate between cumulus clouds that are
forming and those that have reached their maximum size and are
ready to dissipate. Under the former will be found the strong lift
of a "live" thermal current. Under the latter will be little if any
rising air, and possibly strong downcurrents.
Although it takes considerable practice to be able to choose
clouds accurately, there are several indications that are very helpful.
A small cumulus cloud that has just begun to form will have the
typical "cotton-ball" rounded top effect, almost always with a flat
bottom at the condensation level. If watched carefully it will be
seen to boil up slowly as it grows and the top of the cloud lifts
higher and higher. One of the best indications of the fact that such

a cloud is growing are small, wave-like curls that continue to
appear on the sides and top of the cloud. If watched these curls
will be seen to roll over and be swallowed up by the growing
cloud. This manifestation is particularly marked on days of con
siderable wind velocity when the thermals are quite turbulent and
there is a definite rolling movement of the upcurrents forming the
clouds. Clouds that have passed their maximum and are about to
! WICMITA
FAJJ-S
SKETCH MAP OF THE AUTHOR'S 212-MILE GOAL FLIGHT, APRIL 19, 1938
break up are usually characterized by a straggly, ragged appearance!
of their edges.
Many times during the summer months thermal upcurrents anil
the cumulus clouds marking them will be so frequent and evenly
spaced that even a novice pilot will have little difficulty continuing
on his course for many miles. He must remember, however, that
no soaring flight is flown on an absolutely straight line as is .pos
sible with powered flight. There is a continual zigzagging to utilize
the best lift. On some flights made by the author it was necessary

to make a wide tour 10 miles to one side of the course to follow
a line of clouds that looked promising when the air straight ahead
had been quite empty of any clouds.
Sometimes it is advisable to follow a rather wide zigzag course
if there is a possibility of encountering the regular parallel bands
of thermals forming the horizontal rolls which cause what we call
"cloud streets/' Typical cloud streets that appear frequently over
level farm lands can be recognized easily from the ground as long
rows of cumulus clouds practically touching and running parallel
straight downwind. As this condition is not always so easy to rec
ognize in flight, a varied course may be necessary to disclose its
presence. A well developed cloud street is the greatest possible help
on a distance flight, as on one side of the line of clouds there is
practically unbroken lift, making it possible to fly straight for some
time without spiraling and often at fairly high speed.
On the opposite side of continuous lift of a cloud street there is
the down part of the roll which causes a "waterfall" of air as con
tinuous as the rising air on the other side, and of course this should
be avoided. Eric Nessler, writing of his French distance record made
in the spring of 1938, described flying a considerable distance on
the upside of a horizontal roll unmarked by any clouds.
On the author's record distance flight in Texas he twice encoun
tered for about 20 miles a line of clouds so close as to be almost
touching and looking just like a cloud street, but which proved not
to be a continuous horizontal roll as he found downcurrents be
tween each two clouds. However, the areas of lift were so wide and
the downdrafts so narrow that it was possible to fly straight for
some time. The technique used was to fly slowly enough to get
maximum lift under one cloud without spiraling and then fly quite
rapidly through the intervening downcurrent to the next where he
again slowed up and quickly regained whatever height had been
lost.
Strong thermal upcurrents unmarked by any clouds and conse
quently called "dry thermals" are fairly common in certain parts
of the country, especially in the deserts of California, Arizona and
New Mexico. It was with such conditions on a cloudless dav/ that
John Robinson of San Diego set an unofficial altitude record of
10,200 feet in an intermediate sailplane in 1938. Distance soaring
on dry thermals is, of course, harder without the cumulus clouds
as markers to point the way.
A great help occasionally are other indications of lift, such as
leaves, butterflies, and soaring birds. The latter are sometimes very
useful to a soaring pilot. While losing altitude and not knowing

which way to turn he may see a buzzard, a hawk or an eagle spiraling nearby and steadily gaining altitude. Soaring birds do not fear
sailplanes and sometime seem to welcome these big man-made birds
to their element. The author has been able to fly within 10 feet
of a golden eagle when both were almost standing still heading
into a slopewind. On other occasions he has had large hawks fly
straight toward him for some distance to join him in the upcurrent in which they saw him spiraling.
It is interesting that the best modern sailplanes will out-perform
most of the soaring birds, with the probable exception of such
master soarers as the albatross and the man-o'-war bird, in sinking
speed and gliding ratio. However, the birds more than make up for
their deficiencies in this respect by their small size, marvelous
maneuverability and vastly superior experience. A bird will do
more flying in a few months than most pilots in a lifetime; a
soaring pilot can always learn by watching them.
A common indication of thermal action in desert country is
the common "dust devils" seen more rarely in other parts of the
country. These are the cores of thermal columns of rising air.
While on an expedition to Iran the author saw on many mornings
a large number of these small whirling columns of dust rising from
the level plain after ten o'clock in the morning. By noon they
sometimes were visible, because of the dust lifted into the air,
to heights of over 1000 feet. Nearly everyone has seen a small "dust
devil" swirl across a road scattering tumble weeds and dust and
then suddenly disappear or seem to stop. This is the taking off of
a thermal bubble which starts rotating on the ground before it
lifts and starts its upward climb.
In flying straight across country toward a cumulus cloud that is
considerably higher than the altitude at which he is cruising, a
pilot should approach it straight downwind. The reason for this is
that the series of thermal bubbles or the thermal column of rising
air leans with the wind. This fact soon becomes apparent even to
a beginner as he usually finds when he has climbed several thousand
feet in a thermal that he has drifted some distance downwind. The
author once experienced an extreme case of drifting with a thermal
while on a cross-country flight in Oklahoma. Preparing to land into
a high wind when forced down to 350 feet, he encountered a tur
bulent but weak upcurrent. By careful spiraling, he managed to
stay within it to reach a height of only 1500 feet while drifting
over 5 miles with the wind.
While flying cross country for real distance on a day of good
convection, a pilot should usually not be content to climb on ther-

mals lifting him at 5 feet per second or less. He should continue
until he finds lift of 7-10 feet per second registering on his variom
eter. By regaining altitude on strong thermals much time is saved
in utilizing as well as possible the best conditions during the hours
when they exist. While in search of such good lift he should not
become discouraged when encountering strong downcurrents as
these are indications of equally strong upcurrents nearby.
In reaching an area of substantially reduced lift near the top of
a thermal (possibly the bottom or trailing part of a thermal bubble)
after a rapid climb it is better to straighten out and fly downwind
in search of another strong thermal rather than to hang on wasting
time while squeezing a last 200 or 300 feet out of this particular
upcurrent. However, this does not hold true in the late afternoon
when thermal activity is on the decline. Then it is best to get the
most out of any rising air that can be found. Wolf Hirth equipped
his "Moazagotl" with 100 pounds of water ballast to give him speed
while the thermals were strong in the middle of the day. When they
weakened around four or five o'clock he dumped the water and
partly lowered his flaps while flying slowly to reduce his sinking
speed to a minimum and take full advantage of the remaining lift.
If forced low while flying over hilly country it is best to keep a
lookout for ridges faced into the wind permitting slope soaring. It
is also important before dropping low enough to rely on ridge
soaring to check the wind direction by watching cloud shadows,
smoke, or ripple marks on water. Often it is possible to find strong
thermals by watching the terrain only. If there are light-colored
fields in a valley on the windward side of a ridge which has a
U-shaped cut-back or ravine the pilot can be almost sure of finding
strong thermal lift somewhere to the lee of the top of the ravine.
The strongest thermals are those that have risen by convection in
the valleys and been given a boost upwards by the mechanical
action of the wind being deflected over a mountain.
It is often possible to prolong a flight by slope soaring over a
ridge while waiting for a thermal to come along on which to climb
to a height sufficient to make it possible to continue cross country.
Of course, it is not possible to fall back on this type of soaring
when flying over level country, but a careful study of terrain fea
tures will often prevent a premature landing. One should look for
lift over or to the Ice side of light-colored and dry ground and
avoid downcurrents over dense woodland, lakes, and swampy
ground. Strong thermals sometimes form along medium-sized rivers,
a fact the author experienced several times during level country
flights in Texas and Oklahoma.

When forced low enough so that a landing is probable, the pilot
should decide on a landing field while looking for rising currents
to prolong the flight. He may sometimes find it advisable to land
rather than to struggle on in a weak thermal when a stretch of
several miles of unbroken forest lies ahead. Great wooded areas,
such as the Pocono Mountains in Pennsylvania where there are
stretches of 20 miles or more without a clearing large enough to
land in, should not be crossed unless good thermals are abundant
and the pilot is 3000 to 5000 feet high. This area must be crossed
on record flight attempts to the south and southeast from Elmira.
TYPICAL END OF A CROSS-COUNTRY FLIGHT

On a farm in Pennsylvania after take-off at Elmira.
When choosing a place in which to land the pilot should first

make sure that it is large enough. If there are obstructions about
such as high trees, houses, or high tension lines he should keep
them well in mind as he notices the wind direction and makes his
approach. It is a good rule to come in purposely somewhat high
to make sure of clearing all obstructions. Excess height can then
be lost by the the use of spoilers, flaps, and slipping. If the speed is
high when the sailplane is a few feet above the ground and there
is danger of not stopping before reaching the end of the field, the
sailplane should be landed, the control stick pushed full forward,
and the brake pulled on hard. Unless the surface is covered with
ice or wet grass the sailplane can usually be brought to a quick
stop in a surprisingly short distance, and with no danger of over
turning. At the end of the author's long ridge soaring flight he
stopped in 40 feet from a landing at 60 m.p.h., the nose skid dig
ging several inches into the ground without doing any damage.

It is best for the pilot to be prepared to secure the ship in case
of a landing in a high wind when there is no one at hand to assist
him. In Texas the author carried, in addition to wrench, hammer
and pliers to help disassemble the ship, three thin steel stakes with
6 feet of ]4-inch rope tied to them. On one landing in Oklahoma
there was a 30-mile wind blowing and no human help within 2
miles. Holding the nose down with his weight as he climbed out
and pulled the stakes out of the baggage compartment of the
Minimoa, he quickly ducked under the wing and caught the tail
before the nose could rise putting the wings in a lifting position.
Then, balancing the wings, he walked the tail around and swung
the ship until it headed downwind. It was then secured level with
ropes to each wing at aileron hinge fittings and to the nose.
In a case like this it is advisable not to tie the ropes so they are
taut as this may cause undue strain on the wings. If the ship is
left overnight, dew or rain may cause the ropes to shrink and, if not
enough slack is left, this may cause a wing spar to crack. Another
good rule to remember to save the wings from strain is not to hold
a tip rigidly while the sailplane is being towed slowly over uneven
ground. Instead, the man at the wing tip should hold his two hands,
palms toward each other, horizontally about 6 inches apart, allow
ing the oscillation of the tip of the wing to have free play between.
After the landing the pilot should reach in and push the plunger
on his barograph lifting the stylus from the drum. If his is a record
flight he should wait until a responsible witness, preferably a man
of some standing in the community, does this for him and signs the
blank forms carried on the flight. No record attempt should be
made in the United States without these necessary forms obtainable
from The Soaring Society of America, P. O. Box 71, Elmira, New
York. It is also necessary to have an official glider observer repre
senting the S.S.A. and the National Aeronautic Association to start
the barograph and witness the take-off if the flight is for "Silver C"
distance or a national or international record. It is a good idea to
carry two barographs because of the possible failure of one.
Another means of distance soaring is flying crosswind before a
polar or cold front. As explained in Chapter VI, a polar air mass
on coming into contact with warmer air of sufficiently great differ
ence in temperature will cause a line of clouds and storms before
which the air is rising. If a pilot can find such a condition and can
manage to slay out in front of it without being drawn into the
storm he should be able to travel a considerable distance. Line
storms of this nature have been known to extend along a front
over 1000 miles long. This type of distance soaring should be

attempted only by an experienced pilot in a modern, completely
equipped sailplane. Present will be all the hazards of storm flying
described under altitude soaring.
A good idea for cross-country soaring is to have two pilots flying
sailplanes of similar performance to keep within sight of one
another. This not only adds greatly to the enjoyment of the flight
but can be of real help to both pilots. This was proven conclusively
on the first "formation" flight made by Warren Eaton and the
author in the "Falcon" * and the "Albatross II," two Bowlusdu Pont sister ships. In the fall of 1934 they soared together from
Big Meadows to Front Royal, Virginia, a distance of 31 miles. It
was a particularly difficult flight along a serrated part of the Blue
Mountain ridge and was largely possible because it was made with
two ships. There were many times when one flew toward the other
to take advantage of better lift that he had found.
Another thing for the experienced pilot to add enjoyment to his
flight as well as to make a contribution to the art is to carry a pad
and frequently jot down notes as to lift, terrain features, types of
clouds, temperatures at various altitudes, etc. The author usually
has done this with the pad strapped to his right knee with a rubber
band and the pencil slipped under one of the bands when not
being used.
GOAL FLIGHT
Only slightly different from those of an ordinary distance flight
are the considerations and preparations for a goal flight to a defi
nite destination announced to the officials before the take-off. Ac
curate navigation is naturally much more important for this type
of flight. Although Stanley Corcoran made the third longest flight
at that time in America, 202 miles over unfamiliar country and
without any map, the author's 212-mile goal flight would not have
been possible without one. Especially during the latter part of the
flight he had to refer to it frequently as a change in wind direction
and velocity made it necessary to allow for considerable drifting
while flying crosswind.
On a goal flight a pilot stakes his experience and reputation to
make his announced destination; it follows naturally that he should
have a good deal of distance soaring experience before attempting
such a flight. It is a good idea to make it to an airport from which
it will be possible to be airplane towed back to the starting point.
The third category of distance soaring is distance to a preannounced destination with return to the point of departure. This
* Now in the Smithsonian Museum, Washington, D. C.

is made without landing at the turning point, where an official
observer must be stationed. Notable distance and return flights have
been made with the first half into a moderate wind and the return
trip back with the wind. Real record possibilities for this type of
flight exist along the long, continuous ridges in the Allegheny
Mountains where the flight can be made mostly on slope winds
reinforced by thermals while flying crosswind each way.
ALTITUDE SOARING
Altitude soaring ranks on a par with distance soaring with respect
to its value in pilot training, meteorological research and thrill for
the sportsman pilot. However, record attempts involving instru
ment flying in cumulonimbus clouds with possibilities of extremely
turbulent air, lightning, hail, freezing cold and lack of oxygen are
far more hazardous than other types of soaring.
It is necessary to have a strong sailplane stressed to diving speeds
of 200 m.p.h. with good inherent stability and equipped with brake
flaps and a complete set of instruments. These should include, in
addition to those necessary for distance soaring, a good turn and
bank indicator, preferably an electric type operated by dry cell
batteries. The usual type worked by a venturi on the outside of the
fuselage is likely to become inoperative when needed most because
of moisture and ice clogging the venturi. A clinometer is also help
ful. A parachute must be worn on altitude record attempts. Warm
clothing is also essential.
The first mark that a student pilot tries for is the 3280 feet above
point of release for one of the requirements of his "Silver C"
license. This is comparatively simple and requires no more than
the usual thermal soaring technique already described. This altitude
probably will be exceeded by him on his first cross-country flight
which need not and should not involve any cloud flying.
By cloud flying is meant soaring inside clouds by the use of
instruments. It is very helpful and also wise for a student to get
some dual instrument flying training in an airplane or in a Link
Trainer. The latter is a machine widely used for this type of training
as it operates indoors and closely approximates actual flight.
Lacking the facilities or funds first to acquire such training, an
experienced and capable pilot can teach himself instrument flying
quite safely if he is careful. He should first get used to the turn
and bank indicator in continuous spirals. At a good altitude of
several thousand feet, preferably in quiet, clear air, he should keep
the ball centered and the needle of the turn indicator held slightly
to one side as he makes continuous, shallow turns. With his eyes

on the instruments and not looking outside, he should keep the
air speed constant with his control stick. All the time he must
remember that in instrument flying his bodily sensations are of
little or no value to him after the first few minutes of becoming
"blind" or unable to orient himself by visual reference to ground,
horizon, or clouds. He must learn to trust his instruments and not
his feelings.
The first attempts of cloud flying should be in cumulus clouds
of medium size and in which there is little turbulence. After first
flying into such a cloud by being lifted up into it from below or
firing into the side of it higher up, it is a good idea to fly straight
through it or to turn gradually so as to come out soon into clear
/VC. /9+ (/ Hov* t)flu\)
BAROGRAM OF ALTITUDE FLIGHT MADE BY AUTHOR ON

OCTOBER 25, 1938
air. Repeated sorties like this into the cloud will give the pilot
confidence to try more without danger of losing control.
On entering a growing cumulus cloud, the pilot will usually
notice an increase of lift due to the heat liberated by the process
of condensation forming the cloud. He may also find the air within
the cloud to be very turbulent, although sometimes the area of best
lift within a cloud is quite smooth. If it is smooth the pilot should
start a gradual, continuous turn as he watches his turn and bank,
airspeed, clinometer and variometer. It is best to limit this first
spiral to about 5 minutes and then level out and hold a steady
compass course to fly out into clear air. Instrument flying is difficult
and quite a strain on the nerves so it is wise to take it in small
doses at first and gradually work up until it is possible to stay
within a cloud for an hour and longer.
If bad turbulence is encountered at any time the student pilot
should straighten up and fly out of the cloud. If he should get com
pletely mixed up and lose control, which is very easy, he should

open up his brake flaps, or double spoilers, to prevent excessive
speed in a possible dive. Some pilots prefer to put their sailplanes,
particularly if they are not overly strong, into spins to get out of
the cloud without overstraining the wings.
A common fault of pilots attempting altitude flights is that they
do not penetrate deeply enough into a cloud before starting to
spiral. They may be quite content to rise at 5 to 10 feet per second
near the edge of the cloud, not realizing that the lift would be
better than 20 feet per second in the center of the cloud. It is best
to watch the variometer carefully while flying straight into the
cloud heading for its center, and not to start turning until the area
of strongest lift is reached. In most variometers there is some lag
for which it is necessary to allow for best results. The ability to
do this properly can come only from experience.
Before any attempt to set an altitude record or qualify for one
requirement for the "Golden C," 10,000 feet above the point of
release, by cloud flying, a pilot should have plenty of experience in
gaining heights of 500 to 3000 feet inside of cumulus clouds. When
he has this background of experience he should pick with care his
cloud for the high altitude climb. A large, towering, and rapidly
forming cloud is often characterized by slightly curved, flat "cap
clouds" which form over the top of the cumulus which soon pushes
up through them as they slide down over its "shoulders" as others
again form on top and the process is repeated. The most satisfactory
method to reach such a cloud is by airplane towing to it and re
leasing under it. If this type of launching is used it is necessary
to carry a sealed barograph in both airplane and sailplane and
have the airplane pilot an official observer to get credit for the
altitude which counts only above the point of release.
Lift in such a large cloud, which should be utilized before it
reaches the dangerous proportions of a nimbus and becomes a thun
derstorm, may be as high as 20 to 30 feet per second so that, if the
area of strongest lift is reached and stayed in, the climb to 10,000
feet may take only 10 to 15 minutes. During the 1938 German
national contest on the Wasserkuppe several pilots went up into a
thunderstorm and recorded astoundingly powerful vertical currents
of over 60 feet per second. This is 40 m.p.h. straight up! One pilot
was lifted to 20,000 feet in 5 minutes.
In conditions as violent as these there is considerable danger even
for the most experienced pilot. Unless he is equipped with oxygen
tanks he may lose consciousness at such great heights. An equally
serious danger is that of encountering hail which is formed in the
terrific up and down currents of a thunderstorm. This may riddle

the sailplane with holes, greatly diminishing its lift and control
and may also injure the pilot and cause him to lose consciousness.
There have been no cases recorded of a sailplane being hit by
lightning but theoretically there is no reason why one could not
be destroyed by it. Of all these dangers, the one most apt to happen
to the pilot is to have his ship thrown out of control, perhaps on
its back, and fall into a very fast dive and then be broken up by
the sudden and terrific stresses imposed by the violently turbulent
air currents. The soaring pilot who goes up into these conditions
to set records and learn more about the movements of the air is a
hardy pioneer who deserves great credit.
The most important considerations for successful soaring in any
of the categories are careful preparation and constant practice. So
much remains to be learned about this ocean of air in which we
live that, perhaps in this almost more than in any other human
activity, experience remains the best teacher.
CHAPTER X
TRANSPORT GLIDERS
DEVELOPMENT OF PRESENT military importance and

future commercial possibilities is the large transport glider. As used
by the Germans and the Russians these gliders are towed singly or in
trains of from two to twelve behind transport or bombing airplanes.
The troop gliders vary in capacity from one holding six to a
reputed giant capable of carrying forty men.
There are two basic military uses of the troop glider. First is the
air transport of men and supplies from one location to another.
Second is the surprise attack of enemy positions by air-borne shock
troops.
The first time that the latter was known to be used in the war
was during the German invasion of the Lowlands in 1940. Key forts
as well as important bridges in Belgium were captured by troops
landed in gliders. These surprise attacks were probably made by
towing the gliders at night to a point ten to twenty miles from the
objective at sufficient height and with perfect timing so that the
gliders could land, unseen and unheard, in the first faint light
of dawn.
Probably the first time that troop-carrying gliders were used in
any numbers for attack was in the successful air invasion of Crete.
The gliders used in this history-making battle were of 10-12 men
capacity. They were of high wing, monoplane design with a span
of 80 feet, length of 50 feet and overall weight of around 4200
pounds. Construction was similar to general American practice for
sailplanes: wood wings and wrelded steel tubing fuselage, the whole
fabric covered.
A two-wheeled landing gear was used for take-off and then
dropped, the landing being made on the wooden skid. The ratio of
glide was probably about 15 to i and the landing speed around 35-40
m.p.h. These ships were lowed singly or in twos and threes behind
the low wing, all-metal, irimoiored Junkers JU-52 transport planes
at speeds of 100 to 120 m.p.h.
While not replacing parachute troops the glider troops would
seem to have certain definite advantages over them in some respects
196
TRANSPORT GLIDERS
197
and in certain situations. Chief advantage is their ability to land

m surprise attack as compact combat units, together and with their
weapons ready to use, at or near the position where they are most
needed.
Large military gliders can also be used to carry materiel as well
as personnel. Light guns, ammunition, bombs, light vehicles, gaso
line, water, food and medical supplies can be transported in them.
In looking into the practical reasons for the use of gliders for
transport we come to the fundamental truth that you can tow much
more than you can lift. This is the basic economic reason for a
locomotive towing a train of cars, a truck towing trailers, or a tug
towing barges. Comparing gliders with these long-proven methods
of transportation we find that the rails supporting the railroad cars,
the roads supporting the truck trailers and the water supporting the
barges compare with the air supporting the gliders. Similarly, the
retarding friction of steel wheels against rails, rubber tires on roads
and boat hulls through water compares with the aerodynamic drag
of the glider when towed through the air.
As commercial aviation grew out of the technical advances of the
first World War so this very important phase of aerial commerce
will grow out of the use of the troop gliders in the present war. In
analyzing the problem it becomes obvious that carrying all of our
passengers in transport planes is like carrying train passengers in the
locomotive. The future will probably see specially designed loco
motive planes with fuel carried largely in one glider (the "tender")
and passengers, mail and freight in other gliders.
The idea of carrying fuel loads in a glider can quickly become
a practical way of greatly extending the useful range of a bomber or
large transport plane. Transference of the gasoline from the glider
to the bomber can easily be accomplished through a flexible hose
attached to the towing cable. The pilot of the glider can pull up
to a position 30 to 40 feet above the line of flight so that the gasoline
will flow down by gravity to fill the tanks of the airplane.
In considering further the economic reasons for the use of large
towed gliders, both military and commercial, the operating saving
of two engines doing a job now requiring eight or more is imme
diately obvious. Far greater, however, is the saving in equipment
cost. Estimates on production cost of large gliders show that they
will probably cost only one fifth or one fourth as much as transport
planes of comparable capacity. Figuring only three of these gliders
towed behind a transport plane you find that the total equipment
cost of doing a job such as moving an air-borne battalion of infantry
will probably be less than half that of using transport planes alone.

A basic development that will make the launching of trains of
troop gliders as well as future freight and passenger gliders practical
has been successfully tried out experimentally. This is the picking
up of a glider stationary on the ground by an airplane in flight pass
ing overhead. In the tests made a four-seater cabin airplane of 260
horsepower picked up a single-place glider (piloted by the author)
of 5Oo-pound gross weight while flying by overhead at close to 100
miles an hour. The secret of this type of pick-up is a small winch
and brake, in the towplane, which reels out several hundred feet
of line to ease the acceleration of the glider. This proved to be
substantially less than in a shock-cord launching.
Word has come from Russia of the launching of trains of ten or
more gliders behind a single transport plane by picking up the
gliders singly or in pairs at intervals of about four or five seconds.
This was done by placing a winch drum in the nose of each glider
instead of a single drum in the towing airplane.
In considering the problem of getting a train of large gliders into
the air we can suppose that a Douglas DC-3 *s to tow three i4-place
gliders. The gliders will be in an inverted V-formation, each with
its own towline to the towing ship. Let us assume that the necessary
spacing will make a total length of the train about 500 feet.
This means that there will be that much less airport for take-off.
It is possible that the drag of the gliders will lengthen the airplane's
take-off run by 800 feet. This would mean a total of 1300 feet of
runway eliminated. When added to this there is the consideration
of the rate of climb being greatly reduced, the problem begins to
assume sizable proportions and limits the use of such an arrange
ment to the largest airports. Even with the use of powerful winches
to assist such take-offs the results would probably not compare with
those obtainable by the use of the pick-up system with which it will
conceivably be possible to tow a train of gliders out of a field too
small for the safe take-off of the airplane alone.
Let us take a look into the future and imagine commercial opera
tion of an aerial freight or passenger train. For maximum economy
of operation we can think of the "locomotive" and some "cars" of
this "train" flying non-stop from New York to San Francisco at an
average cruising speed of around 200 m.p.h. A few miles east of the
Akron, Ohio, airport during the letdown the captain of the train,
in the towplane, after checking with the control tower at the airport,
tells the pilot of one of the gliders through intercommunication
radio to release. This glider then drops off and slows to its most
efficient glide of about 70 m.p.h. The train continues on and, slow
ing to about 120 m.p.h., swoops low over the pick-up station on the
TRANSPORT GLIDERS
199
airport and hooks on the towline of another glider which becomes

part of the train. Shortly afterwards the released glider lands
smoothly and rolls up to the station. With a probable towing ar
rangement of each glider having its own towline to the towplane,
this process is repeated in variations of several gliders released or
picked up at a time as the train continues.
Gliding and soaring, the finest sport developed by man, has made
many practical contributions to aviation in the fields of pilot train
ing, aircraft design and meteorology. It now has made possible the
development of the large transport glider as an important military
weapon. Its greatest contribution to human society, low cost, com
mercial aerial transport, is bound to follow.
APPENDIX
GLIDING AND SOARING LICENSES
LICENSE PINS for MOTORLESS FLIGHT
"A"
"B"
SLIDING
"C"
"Silver C"
SOARING
"A"Two gliding flights of at least i minute duration, "S" turn,

normal landing.
"B"Two gliding flights with 360 turns, one to right and one to
left, landing so as to come to a stop within 100 feet of a desig
nated mark.
"C"One soaring flight in which an altitude greater than that at
the starting or releasing point is maintained for at least 5
minutes.
"Silver C"Two or three soaring flights in which the three fol
lowing requirements are made:
(i) Distance, 32 miles (50 km.); (2) Altitude, 3280 feet (1000
m.); (3) Duration, 5 hours.
"Golden C"(i) Possession of "Silver C" license; (2) Distance,
185 miles (300 km.); (3) Altitude, 10,000 feet (3000 m.).
These licenses are awarded by the Soaring Society of America,
Inc., P.O. Box 71, Elmira, N. Y., and the qualifying flights for them
must be witnessed by an official Glider Observer. A sealed baro
graph must be carried on qualifying flights for Silver C and Golden
C licenses with the possible exception of the 5-hour duration, pro
vided it is made within sight of the Observer.
CIVIL CERTIFICATES
The Civil Aeronautics Administration of the United States Gov
ernment issues to pilots of motorless aircraft, licenses, known as
certificates of competency, which are now required by law in many
states. The three grades are as follows:
Student Glider Pilot Rating
Private Glider Pilot Rating
Commercial Glider Pilot Rating
Requirements for these ratings are listed in the Civil Air Regula
tions which are obtainable from the nearest C.A.A. Inspector or
from the C.A.A. headquarters in Washington, D. C.
201
s
3.3
"
s O '.3 !?S SJ>!E O.M C S tj 2 ^

O<<<
tOO<
-t.2M
5, <
S,
-Cl
(M
h
3
e
I
i I !
* 8
i S
H
22
01
..;
r? - .
r?
r^,^
S 'a
'3
xA
c/
'^v
&
^
o
^
c 2 ^
3 ^S
"5S
COCO
.w?Bîo5
x^2S
0
rt
-CÔ5-
'->'->x~ve>;f
K
V.
<0
'"
*i
r^ _ r >;
P.
HiJH^^^lflfWllKKl!
g 18.
fc o> g -1 *S -| -S *1 =|| *1 1 #
I H^lff
5
J
-H
>^
S
rI
a3 p ti o ^^ ^s ^s aw tj s - z 3C ^ y o ûi ->* f o
-0<<<ÔO<Q-3
w'ii
pi-
r^
u w
i-i s>
On 1^
3 "
3"
ST 1"1
O 1"1
U 1^-
3"
OJ^-^"
3 " "3
APPENDIX 203
- i
.
.-
P
~
*fl
C
r+
g
*->
rji
<O
O
S o S ?? S3 . SU
S'SS'^S'IS'S
*1
* Sis 1*1
">
'>
>
4_>
r^
CO
SI SP =
?
. 0 .
crt i-i C
. C . cu
oo w ^ N
- t: <M c
.03
. CU
60
S ^ffi
3^3^
f 3 S.5 fa fcS
en
*-"
U-l
*'
CO
g$
S-
15
*:
CO
<
<
J
M-4
~s
*>
i>~
<j o
^j 
. "
*k o
.(-).(-!
. *
<^
K.^Zs ^^;
. c
- <u oo <u
-< ^5
en
&
_
JO 4_T
oi ;! 01 N
1 S3
S
3
r- <M oo O)
co oi co co
O)
~ S
CM
i '
'
CM
0,Z
<
1Z
P4
4-1
r*
- S
*S
c*
A
^"^
<
s
PQ
5o
!H:
"a
ooCO
o
-/:
~~*
&J3
00^3
CM
^*-OO
C5
" P
P-"
<-" S
a,T3

Cd
F"4
y,^^
fid
;"
^
*"
10
c/5
CO
^"^
'""^
^5
"
xA
t-*S
co 3
i>^
co 3
""
w 3 >^H-I *j J? >,-H >^.ii: >N^;J

CL.O
C
S
g
?
en
-:

0
CJ
en
K-;
?^
P<
c?
.S
SI S o
s-<
. o3 ey
^pq
c3
OH
CM
faS! a Isi ^ffLffa-jil--
g
^
co ~ o to
Oi ft X CM
^ cu o>
bo - .
sg
w
^"8
.j '
CM
-
S 0> g xft g
M
S
CM X
coO
hi-J
<
CM c^, m
- o -
r* co
"
CO '4~*
^2 S
^
0)
e
>'
W
_44 *-~^
^T1 ^
CO r
->
Q)
O> o
- <M
iT>
^ ^ o ^ w g ^^ QJ in
eo^-St-^vS.^-^^
co 03 W
co o3 W -^^ * ^
^J
<
+4
rco
-
3&SJO,
Si
o
?s ?g
^"Sî
. 03
. CU
bO^bCW
3 "^
3 "^
U.S.A.
Jan. 1903
Jan. 1903
Wright (U.S.A.), 1.1 min.
Wright, 1.1 min.
Oct., 1911
Oct., 1911
Wright (U.S.A.), 9.7 min.
Wright, 9.7 min.
Aug. 30, 1921
Klemperer (G.), 13 min.
Feb. 7, 1923
Dcscamps (F.), 1774 ft.
July 26, 1925
Auger (F.), 2240 ft.
Aug. 8, 1928
Dittmar (G.), 2530 ft.
Apr. 25, 1929
Nehring (G.), 3868 ft.
July 20, 1929
Kronfeld (G.), 6482 ft.
July, 1932
July 30, 1929
O'Meara, 5370 ft.
Kronfeld (G.), 8320 ft.
Feb. 17, 1934
June 30, 1934
Dittmar (G.), 14,137 ft.
duPont, 6223 ft.
May 22, 1937
July 5, 1938
Steinig (G.), 18,720 ft.
duPont, 6804 ft.
Aug. 5, 1938
Dreschel (G.), 21,398 ft.
Nov. 11, 1938
Ziller (G.), 22,434 ft.
July 4, 1939
Stanley, 17,264 ft.
International
DURATION (SINGLE PLACE)
June 5, 1938
Wills, 10,080 ft.
Great Britain
France
Sept. 17, 1934

Nessler, 3267 ft.
May 25, 1935
Roger, 5161 ft.
March 21, 1938
Lamort, 5926 ft.
June 30, 1938
Nessler, 11,532 ft.
Feb. 7, 1923
Descamps, 1774 ft.
July 26, 1925
Auger, 2240 ft.
ALTITUDE (SINGLE PLACE)continued
Aug. 30, 1921

Klemperer, 13 min.
Aug. 8, 1928
Dittmar, 2530 ft.
Apr. 25, 1929
Nehring, 3868 ft.
July 20, 1929
Kronfeld, 6482 ft.
July 30, 1929
Kronfeld, 8320 ft.
Feb. 17, 1934
Dittmar, 14,137 ft.
May 22, 1937
Steinig, 18,720 ft.
Aug. 5, 1938
Dreschel, 21,398 ft.
Nov. 11, 1938
Ziller, 22,434 ft.
Germany
O
d
0
I
hO
O
APPENDIX 205
CO
oo co
00 *>
CO .
o ~
o
-* J-l
rt . rt
05
r~
N!
oo
. X
CO 3
JZ
>^ u
rt en
00
05 c
00 CO
- rt
<
^ 2 ^ rt
bo
3 ^
- 'S r^. 2
CO
^_-
CO
?~
** 
01
oo
- u

C5 O
"*
!-i
?^^
^ î co j
-- - en oo
O
C5 ^-^
r-C
'oo" 5x "" -
'.X ^"^> cw..
rz^ - aj o
H
H d
^ - r- .> C >-7T. ^ ~ N
U.S.A.
May 3, 1937
Spilger-Hahenne (G.),
7,283 ft.
Sept. 18, 1937
Ziller-Quadfasel (G.),
10,840 ft.
Aug. 12, 1940
Barringer-Froehlich,
14,960 ft.
July 23, 1939

Kartacheff-Chechoulkine (U.S.S.R), 212 mi.
219
mi.
mi.
mi.
Great Britain
(TWO PLACE) *
July 4, 1939
Barringer-Rodenburg,
6558 ft.
Aug. 12, 1940
Barringer-Froehlich,
14,960 ft.
June 28, 1936

Slatter-Batterson,5967 ft.
ALTITUDE (TWO PLACE)
DISTANCE AND RETURN (TWO PLACE)
Oct. 21, 1936

July 28, 1936
Ilchencho-Loguin
Slatter-Batterson, 25
(U.S.S.R.), 83 mi.
Apr. 12, 1937
Kneis-Beck (G.), 121 mi.
May 27, 1937
Ilchencho-Emerik
(U.S.S.R.), 253 mi.
May 15, 1938
June 29, 1938
Ilchencho-Zelenkova
Smith-de Medo, 28
(U.S.S.R.), 345 mi.
July 17, 1938
April, 1939
Kartacheff-Savtzov
Essery-Hepburn, 43
(U.S.S.R.), 385 mi.
July 13, 1940
Stanley-Schweizer,
mi.
International
DISTANCE
France
May 3, 1937
Spilger-Hahenne, 7283
ft.
Sept. 18, 1937
Ziller-Quadfasel, 10,840
ft.
Apr. 12, 1937

Kneis and Beck, 121 mi.
Germany
O
d
0
I
July 4, 1936
Slattcr-Buxton, 8:48
DISTANCE AND RETURN (TWO PLACE)
July 9-10, 1938

Murray-Sproule, 22:13
ly. !937
Fox-Murray
9:48
Two-place records were not officially accepted by the F.A.I, until 1936.
June 14, 1938

Kartacher-Naoumor
(U.S.S.R.), 12 mi.
Aug. 10, 1938
Huth-Biandt (G.), 186
mi.
Fox-Murray
(G.B.), 9:48
Nov. 26, 1937
Jachtmann-Klossdork
(G.)> 13:59
Apr. 9, 1938
Makaroff-Godovikoff
(U.S.S.R.), 19:08
Sept. 5-6, 1938
Kahlbacker-Tauschegg
(G.), 23:41
Sept. 9-10, 1938
Fuhringer-Kahlbacker
(G.), 40:38
Dec. 9-11, 1938
Boedecker-Zander (G.),
50:26
1937
DURATION (TWO PLACE) *
Aug. 10, 1938

Huth-Brandt, 186 mi.
Sept. 5-6, 1938

Kahlbacker-Tauschegg,
23:41
Sept. 9-10, 1938
Fuhringer-Kahlbacker,
40:38
Dec. 9-11, 1938
Boedecker-Zander, 50:26
Nov. 26-27, 1937

Jachtmann-Klossdork,
hO
O
X
m
~ZL
r
8
t
oo
r^"
<
00
CO
05
o ? _e
<o
00
V.
F*
2
53
oo
00 00
$2
O5 Sr>
*"
ki
1H
-<
OO
52
O5 ^
1-H
^ -U
OO"
8
(M
O5
FN
^ T3
OO"
C^
F-l
&
5-< J?
"
'
<M '
^03
C4
_ TI
a.0
<
"
^
<
g
H
2
H
fe
j
[_
8s
u -2
>J
>j
^3
1^1
QPPP
>
5
^J
<
<
'-i
r1
"^
oo
0*
^
03
10
>^
OO
co ti
05 C
~ o
. PLl

>Lx 05
S 
-
O c/D
rt
APPENDIX 209
AMERICAN "SILVER C" PILOTS
Interna
Interna
U.S. No.
Name
.S.No.
tional No.
Name
tional No.
U
1
J. K. O'Meara
Elmer Zook
12
20
2
Richard C. duPont
Harland McHenry
21
32
Lewin B. Barringer 65
Harvey Stephens
22
3
Stanley W. Smith
Donald Stevens
236
23
4
Emil A. Lehecka
Parker Leonard
24
237
5
6
Henry N. Wightman 238
Woodbridge P. Brown
25
Emerson Mehlhose 239
Randall M. Chapman
26
7
8
Chester J. Decker
Alan R. Essery
27
240
Harland C. Ross
510
Richard Johnson
28
9
10
Arthur B. Schultz
5H
Joseph Steinhauser
29
11
Warren Merboth
871
Harold W. Huber
30
12
Robert Stanley
872
L. Howard Morrison
31
13
Stanley Corcoran
873
Rayman H. Parker
32
14
Theodore Bellak
74
Wm.
Horace
Putnam
33
15
Robert Auburn
875
Lyle
Allan
Maxey
34
16
Julian Hadley
876
Frank S. Boggs, Jr.
35
17
Floyd Sweet
877
Henry Stiglmeier
36
__*
18
John Robinson
Frederick
R.
Dent,
Jr.
37
'9
Udo Fischer
AMIRICAN "Goi.m N C" PILOTS
Robert Stanley
*
3
John Robinson
Chester Decker
4
Lewin B. Barringer
BRITISH "SILVER C" PILOTS
(British number before each pilot's name, and the international number after it.)
1 G. E. Collins
26
26 E. E. H. Collins
59}
2 P. A. Wills
595
45 27 J. L. Wordsworth
3 R. G. Robertson
621
75 28 Mrs. J. Price
4 S. Humphries
85 29 G. M. Thompson
622
5 J. C. Neilan
174 30 L. R. Robertson
625
E. Thomas
6 C. Nicholson
177
856
7 Miss N. Heron-Maxwell
208
857
32 I. Pasold
241 33 H. Tudor Edmunds
8 P. M. Watt
858
9 H. C. Bergel
859
244 34 J. C. Dent
291 35 L. H. Barker
10 A. L. Slater
860
861
298
n G. O. Smith
36 D. F. Greig
1004
12 J. S. Fox
338
37 A. J. Deane-Drummond
A.
Ivanoff
1005
13 R. S. Rattray
542
38
1006
14 P. B. N. Davis
39 A. W. Lacey
543
M.
H.
Maufe
1007
40
15 G. H. Stephenson
545
J.
Parker
1008
41
560
16 D. G. O. Hiscox
E.
H.
Taylor
1009
42
561
17 K. G. Wilkinson
1061
562
18 J. E. Simpson
43 K. M. Chirgwin
1062
563
19 J. V. Rushton
44 R. Pasold
1063
20 G. A. Little
564
45 J. W. S. Pringle
1064
21 K. Lingford
565
46 J. A. Rooper
1093
566 47 J. H. Saffery
22 J. S. Sproule
1094
567
48 P. M. Thomas
23 K. W. Turner
1095
568 49 G. L. Raphael
24 E. J. Furlong
1096
585
5 A. Davies
25 S. C. O'Grady
BRITISH "GOLDEN C" PILOTS
i Philip Wills
4
* International numbers not yet assigned due to war delays.
CLUBS
GLIDING CLUBS AND SCHOOLS IN THE UNITED STATES
Alabama
University
Arizona
Phoenix
California
Alameda
Claremont
Eureka
Glendale
Laws
Los Angeles
Modesto
Pasadena
San Diego
San Francisco
Van Nuys
Colorado
Denver
Connecticut
East Hartford
Waterbury
Delaware
Newark
VVilmington
Florida
Daytona Beach
Gainesville
Perry
Georgia
Atlanta
Idaho
Wesleyan
University of Alabama Glider Club

Southwest Soaring Club, Phoenix Junior College
Glider Club, 1603 Paru Street
Adiabats
Humboldt Flying Club, 523 T Street
Aero Soaring Group, 606 W. Vine Street
Inyo-Mono Soaring Association, Inc.
Southern California Soaring Ass'n, Inc., 738
South Bristol Ave., W. Los Angeles
Modesto Glider Club, 138 Waterford Road
California Institute of Technology Aero Club
Associated Glider Clubs of Southern California,
4141 El Cajon Avenue
Soaring Society of Northern California, 406 Sutter Street
The Briegleb Soaring School, 16005 Bassett Street
Broadmoor Sport 8c Training Camp, 1754 Broad
way
Cactus Glider Club, 1543 Kearney Street
Clark Glider School, Box 76
Waterbury Glider Club, 172 Hamilton Avenue
Delaware College Soaring Club, Evans Hall, Uni
versity of Delaware
Delaware Soaring Society, 1808 Sycamore Street
Model Airplane & Glider Club, Rt. i, Box 218
Seagull Glider Club, University of Florida
Perry Aviation Club
Atlanta Aero Engineers, 2049 Robson Place, NE
The Gooding College Glider Club
210
APPENDIX 2
Illinois
Chicago
Glen Ellyn
LaGrange
Lockport
Chicago Gliding Club, 7828 So. Aberdeen Ave

nue
Chicago Junior Chamber of Commerce, c/o Mr.
Ray Donlan, i No. LaSalle Street
Gage Park Glider Club, 6919 Yale Avenue
Lane Glider Club, c/o Mr. C. V. Olson, 8812
Union Avenue
Motorless Flight Institute, 1641 Addison Street
Soaring and Gliding Club of Chicago, 2207
School Street
Soaring Club of Chicago, c/o Mr. Lincoln Har
ris, 135 So. LaSalle Street
Silver Hawk Glider Club, 806 Crescent Blvd.
Wolf Glider Club, 212 Fourth Avenue
Chicagoland Glider Council, Inc., 426 Spring
Avenue
Frankfort Sailplane Company
Lewis School of Soaring, Lewis-Lockport Air
port
Indiana
Angola
Attica
Bremen
La Porte
Notre Dame
South Bend
West Lafayette
Tri-State Glider Club, 319 S. Darling Street

Attica Glider Club, West Mill Street
Bremen Glider Club
La Porte Glider Club
Notre Dame Aeronautical Society
South Bend Glider Club, 1155 N. Meade St.
Purdue Glider Club, Purdue University Airport
Iowa
Iowa City
Aeronautical Club of the University of Iowa
Kansas
Manhattan
Topeka
Wichita
Kansas State Glider Club, Kansas State College

Topeka Soaring Club, 209 Kansas Ave.
University of Wichita Glider Club
Louisiana
Baton Rouge
New Orleans
Louisiana State University Glider Club, Box 809

Tulane Glider Club, 1831 Bordeaux St.
Maryland
Riverdale
Glider Club of Engineering & Research Corp.
Massachusetts
Boston
Cambridge
Wayland
Worcester
Northeastern University Soaring Society

Aeronautical Engineering Society, Mass. Institute
of Technology
Harvard Gliding Club
M. I. T. Glider Club
Altosaurus Club, c/o Alan Bemis
Worcester Polytechnical Institute Glider Club

Michigan
Ann Arbor
Berkley
Detroit
Suttons Bay
YVyandotte
University of Michigan Glider Club

ABC Glider Club, 3184 Tyler Ave.
Detroit Glider Council, 15100 Woodward Ave.
Lawrence Tech Soaring Society, 15100 Wood
ward Ave.
University of Detroit Glider Club, 16621 Prairie
Frankfort Soaring Association
Four Ace Glider Club, 151 Wabash St.
Depoy Motor Company Glider Club, 108 Syca
more St.
Suttons Bay Glider Club
Wyandotte Gliding Club, 367 Oak St.
Minnesota
Minneapolis
University of Minnesota Flying Club
Missouri
Robertson
St. Louis
Curtiss Wright Airplane Co. Glider Club

St. Louis Soaring Ass'n., No. i Fordyce Lane
Montana
Conrad
Conrad Glider Club
Nebraska
Lincoln
Omaha
University of Nebraska Glider Club

Omaha Aero Club, 4971 Miami St.
New Hampshire
Durham
University of New Hampshire Flying Club
Frankfort
Milan
Niles
New Jersey
Bayonne
Glen Rock
Hillsdale
Interlaken
Morris Plains
Newark
North Haledon
Wycoff
New York
Albany
Buffalo
Ehnira
Ithaca
Bayonne Technical High School Glider Club

North Jersey Soaring Ass'n, 34 Cambridge Place
Pascack Valley Gliding Association, 341 Washing
ton Ave.
Atlantic Seagulls Soaring Club
Morristown Glider Club, 2 Sherman Ave.
Associated Glider Clubs of New Jersey, 91 Halsey St.
Newark Glider Club, P. O. Box 134
Y Flying Club, 395 Sussex Ave.
Glider Club, c/o Albert Boyd, High Mtn. Rd.
Ace Gliding Club, c/o J. Erlenbach, Jr.
Capital Aviation Society, 17 Steuben St.
Buffalo Glider Club, 68 San ford Ave.
Curtis Gliding 8c Soaring Ass'n., Engineering
Dept., Bell Aircraft Co.
Elmira Area Soaring Corp., Federation Bldg.
Elmira Gliding Club, Federation Bldg.
Ithaca Gliding Club, Ithaca Airport
APPENDIX 213
Kingston
Maine
Middletown
New York City
Rochester
Ohio
Akron
Cincinnati
Cleveland
Crestline
Dayton
Gahanna
Ironton
Kent
South Euclid
Wauseon
Xenia
Oklahoma
Stillwater
Glider Club, 130 Pearl St.

Nanticoke Valley Soaring Society, 20 Main St.
Middletown Glider Club, 82 East Ave.
Airhoppers Gliding 8c Soaring Club, c/o A. Dawydoft, 79 7th Ave.
Eastern States Soaring Association, c/o C. Gale,
The Sportsman Pilot, 515 Madison Ave.
Rochester Glider Club, 43 Manchester St.
Akron Advanced Flying Club, 300 E. Exchange
St.
Akron Glider Council, 277 Brown St.
Albatross Birdmen, 818 Wade St.
University of Cincinnati Glider Club
Case Aero Club, Case School of Applied Science
C.T.S. Glider Club, Cleveland Trade School
Crestline Glider Club
Dayton Glider Club, 202 Virginia Ave.
Gahanna Glider Club
Valley Glider Club, c/o C. M. Hardy, Tri-State
Pattern Works
Kent State University Glider Club, 615 Park Ave.
Beaconwood Glider Club, 1720 Beaconwood Ave.
Civilian Air Reserve
Xenia Glider Club, 310 W. 3 St.
Wa tonga
Oklahoma A. 8c M. Glider Club, Oklahoma

Agricultural and Mechanical College
Cloud Buster Club, c/o Tom Oler
Oregon
Junction City
Junction City Glider Club
Pennsylvania
Aspinwall
Chester
Intercourse
Oakmont
Philadelphia
State College
Sykesville
Glider Club of Aspinwall High School

Penna. Military College Glider Club
Intercourse Glider Club
Carnegie Tech Glider Club
Phi Kappa Sigma Glider Club, 3539 Locust St.
Roxborough Aero Club, Kendrick Recreation
Center
Falcon Glider Club, 97 S. i8th St.
Glider Club, 10 Virginia Ave.
Glider Club of the Boys' Club, 4114 Penn Ave.
Shadyside Academy, Box 7374 Oakland Sta.
Penn State Aero Club, 135 So. Frazier St.
Sykesville Gliding 8c Soaring Club, Box 195
Tennessee
Chattanooga
Chattanooga Glider Club
Pittsburgh

Texas
Dallas
Denton
San Antonio
Utah
Roosevelt
Salt Lake City
Virginia
Charlottesville
Front Royal
Washington
Pullman
Seattle
West Virginia
Wheeling
Wisconsin
Ellsworth
Neenah
Dallas Glider Club, 4516 Fairfax Ave.

Glider Club, 405 Bernard St.
Aero Club of San Antonio Vocational 8c Tech.
School
Roosevelt Glider Club, Box 172
Salt Lake City Glider Club, 2376 S. 8 St., East
University of Virginia Glider Club, Engineering
School
Glider Club, c/o Seddon Nelson, American Vis
cose Co.
State College of Washington Aero Club
Seattle Glider Council, c/o Amos Wood, 2659
47 St., NW
Wheeling Aero Club, c/o K. Halpny, 7 8c Market
Sts.
Glider Club
Glider Club, 414 Oak St.
BRITISH GLIDING CLUBS
Compiled by the British Gliding Association,

IIQ, Piccadilly, London, II7 ./
Beacon Hill Gliding Club, W. P. Harris, Sec., 22 Hamlet Road,
Southend, Essex.
Bristol Gliding Club, H. H. Maufe, Hambrook House, Hambrook.
Cambridge University Gliding Club, J. W, S. Pringle, i Benet
Street, Cambridge. Flying ground at Caxton Gibbett.
Channel Gliding Club, F. G. Whitnall, 16 High Street, Cheriton,
Folkestone. Hangar at Arpinge.
Cornwall Gliding Club, J. W. Graham, Red House, Tywardreath.
Flying ground at Rosenannon Downs.
Cotswold Gliding Club, J. D. Pether, Culver's Close, Burford, Oxon.
Training at Minster Lonell.
Croydon Gliding Club, N. V. Marshall, Hollydena, West Hill,
Epsom.
Derbyshire and Lancashire Gliding Club, C. Kaye, 63 Clarkhouse
Road, Sheffield. Headquarters at Camphill, Great Hucklow.
Devon Gliding Club, S. G. Tolman, Journal Office, Exmouth.
Dorset Gliding Club, L. A. Lansdowne, The Portman Arms Hotel,
East Chinnock, Leovil, Somerset. Flying at Maiden Newton
and Kimmeridge.
East Grinstead Gliding Club, G. J. Smith, "Tolskity," Sackville
Lane, East Grinstead, Sussex.
APPENDIX 215
Essex Gliding Club, W. \Vebster, 113, Coombes Road, Dagenham.
Furness Gliding Club, J. S. Redshaw, 18, Fail field Lane, Barrow-inFurness, Lancashire. Soaring sites at Moorside and Bootle Fell,
Cumberland.
Harrogate Gliding Club, E. T. W. Addyman, The White House,
Starbeek, Harrogate.
Hull Gliding Club, R. E. Havercraft, 216, Park Avenue, HU1. Fly
ing ground at Hendon Aerodrome.
Imperial College Gliding Club, Imperial College of Science, South
Kensington, London, S.W. 7. Flying at Dunstable Downs.
Kent Gliding Club, Miss R. H. Sinclair, Lade Place, Sutton Courtenay, Berkshire. Training ground at Lenham.
London Gliding Club, Tring Road, Dunstable, Bedfordshire. Club
house, hangar, flying ground at Dunstable.
Midland Gliding Club, M. F. Barnes, 100, Holly Road, Birming
ham 20. Soaring site at Long Mynd, 3 mi. WSW of Shurch
Stretton, Salop.
Newcastle Gliding Club, A. P. Miller, 25, Home Avenue, Walkerville, Newcastle-on-Tyne, 6. Soaring site at Chillingham.
Norfolk and Norwich Aero Club, Gliding Section, J. F. Taunton,
Municipal Aerodrome, Norwich.
Oxford University and City Gliding Club, Mrs. H. Aspell, 5, Holywell, Oxford.
Portsmouth and South Harts Gliding Club, R. G. H. Parnell, 128,
New Road, Portsmouth. Flying ground at Portsdown Hill.
Southdown Gliding Club, A. York Bramble, 7 a, First Avenue, Hove
3, Sussex. Flying grounds at Devil's Dyke, Brighton.
Yorkshire Gliding Club, L. A. Alderson, 32, Wensley Green, Chapel
Allerton, Leeds 7. Flying ground at Sutton Bank.
SCOTLAND
Dumbartonshire Gliding Club, J. V. Campbell, Kirklea, Cardross
Road, Dumbarton.
Inverness Gliding Club, F. Oliver, 13, Leys Drive, Inverness.
Perth Gliding Club, R. Mackelvie, View Cottage, Union Road,
Scone, Perthshire.
Scottish Gliding Union, J. W. Gardner, Journal Office, Alloa.
Soaring site at Lomond Hills, Fifeshire.
NORTHERN IRELAND
Ulster Gliding Club, N. P. Metcalfe, c/o Ulster Spinning Co., Ltd.,
Belfast. Flying ground at Downhill, Magilligan Strand, Lon
donderry.
CHANNEL ISLANDS
Jersey Gliding Club, A. J. Scriven, Quainton, Samares, Jersey. Fly
ing ground at north end, St. Quen ;s Bay.
WALES
Swansea and District Gliding Club, A. H. Knott, 209 a, High Street,
Swansea.

BIBLIOGRAPHY
(This list does not include all the books on the subject, but only the
best and most useful for the student of motorless flight.)
American Books
Gliders and Gliding, Lt. Com. Ralph S. Barnaby. Ronald Press,
New York, 1930.
Gliding and Soaring Manual. (Official manual of The Soaring
Society of America.) Stone Aircraft Co., Box 57, Detroit, Midi.,
English Books
Gliding and Sailplaning: A Beginner's Handbook, F. Stamer and
A. Lippisch (translated from German). John Lane, The Bodley
Head, Ltd., London, 1930.
Sailplanes: Their Design, Construction and Pilotage, C. H. Latimer
Needham. Chapman and Hall, Ltd., London, 1937.
Kronfeld on Gliding and Soaring, Robert Kronfeld. John Hamilton,
Ltd., London, 1932.
German Books
The Art of Soaring Flight, Wolf Hirth (translated from German).
Stuttgarter Vereinsbuchdruckerei AG., Stuttgart, 1938.
Handbuch des Segelflie gens, Wolf Hirth. Franch'sche Verlagshandlung, Stuttgart, 1938.
Die Praxis des Leistungs-Ser gel flie'gens, Erich Bachem. C. J. E. Volckmann Nachf., Berlin-Charlottenburg, 1936.
Periodicals
Soaring (monthly). The Soaring Society of America, Inc., P.O. Box
71, Elmira, New York.
The Air Bubble (occasional, mimeographed). Chicagoland Glider
Council, Inc., 426 N. Spring Ave., La Grange, 111.
The Thermal (occasional, mimeographed). Southern California
Soaring Ass'n, 738 So. Bristol St., Los Angeles, Calif.
M.F.I. News Bulletin (occasional, mimeographed). Motorless Flight
Institute, Gliderport-Chicago, Chicago Heights, 111.
The Sailplane and Glider (monthly) . H. O. Davies, 13, Victoria St.,
London, S.W.i, England.
Flygtidningen. Sallerupsvagen 26a, Malmo, Sweden.
Amicale du Vol a Voile Frangais (monthly). Aero Club de France,
6, rue Galilee, Paris, France.
Flugsport (bi-monthly). Hindenburgplatz 8, Frankfurt a.M., Ger
many.
Samolet (monthly). Moscow, U.S.S.R.
APPENDIX 217
GLOSSARY
AerobaticAcrobatic maneuvers in flight, such as loops, rolls, etc.
AileronMovable surface for lateral control, forming part of outer,
rear portion of wing.
AirfoilCross-section design of wing or lifting surface.
Angle of attackAngle between chord line of wing and level line
of flight of fuselage.
Aspect ratioRatio of span to mean chord of wing.
BankLateral inclination of glider while turning in flight.
BurbleAn eddy in an airflow.
CabaneA structure on top of glider fuselage forming support for
wings.
CantileverA wing having its supporting structure within itself.
CentroidA geometric property of a beam, spar member, etc., used
in strength calculations. It is the center of area.
ChordThe width of a wing measured from leading edge to trailing
edge.
Crack-upAn accident involving serious structural damage to a
glider.
Dep controlControl stick with wheel for lateral control. Named
for Deperdussin, the Frenchman who invented it.
DihedralAngle between horizontal line drawn laterally from cen
ter of bottom side of wing and line formed by the wing inclined
upwards.
DopeA nitro-cellulose fluid used to tighten and protect fabriccovered surfaces of a glider.
D-tubeStructure formed by combination of spar and curved metal
or plywood surfaces of a glider.
DragRetarding force due to aerodynamic resistance.
ElevatorHinged, horizontal tail surface for controlling glider in
the vertical plane.
EmpennageThe tail of the glider, including horizontal and verti
cal stabilizing and control surfaces.
FairingStructure added to glider for streamlining or lessening of
drag.
FairleadsLeading of control cable to prevent cutting or chafing.
FinFixed, vertical tail surface for longitudinal stability.
FishtailA sideways, level, alternately skidding maneuver to de
crease excessive flying speed close to the ground.
FlapsMovable surfaces hinged to rear portion of wing to facilitate
approaches to landings by increasing lift and drag.
FuselageThe body of the glider.
Gliding ratioRatio of distance covered horizontally to height lost
vertically.
Gross weightWeight of glider fully loaded with pilot and equip
ment.
GussetA triangular piece of plywood or metal fastened across a
structural joint to increase stiffness and strength.
n -Yixed projection on control surface providing leverage action
for control cable or tube.

IdlerAn idler horn, pushrod, gear, etc., is a member of a system
used to change direction, rotation, or to facilitate connections.
JigA fixed form or mold to facilitate exact fabrication and duplica
tion of parts.
LongeronPrincipal longitudinal structural members of a fuselage.
MonocoqueA type of fuselage construction the strength of which
is largely in a metal or plywood shell due to its round or oval
design.
NoseThe front of the glider's fuselage.
PlacardProminent notice painted on side of glider fuselage.
Radio-sondeA radio meteorgraph; an instrument sent aloft (50,000
to 100,000 feet) on a balloon, sending out continuous signals
of soundings of temperature, altitude and relative humidity
which are picked up and automatically plotted on paper by a
machine on the ground.
ReleaseMechanism in nose of glider holding ring at the end of
towline until tripped by pilot.
RudderVertical, hinged tail surface for directional control in
horizontal plane.
Sinking speedThe rate of vertical descent while gliding.
Skid (noun)Curved, wooden runner under fuselage forming part
of landing gear.
Skid (verb)A sideways, slipping maneuver resulting from overcontrol of the rudder.
SlipA maneuver where the glider loses height sideways in the
direction of the down wing while banked.
SpanThe distance between the wing tips of a glider.
SparBeam used as principal structural unit of wing.
SpoilerSmall, hinged control surface on upper side of wing to
decrease lift and increase drag to steepen glide path in landing.
StabilizerFixed, horizontal tail surface for stability in vertical
plane.
StallA loss of lift due to insufficient air speed over the wing.
StringerA light, longitudinal structural member to stiffen a fuse
lage.
StrutA bar or rod used as an outside structural member of a glider.
TailplaneHorizontal tail surface including stabilizer and elevators.
TorqueA twisting or rotary force.
TowlineRope, cable, or wire used to tow a glider.
TurnbuckleA coupling with internal screw threads for regulating
the tension of a wire or cable.
WashoutDecreased angle of attack of wings toward tips.
ABBREVIATIONS
B.G.A.British Gliding Association
C.A.A.Civil Aeronautics Administration (U.S.A.)
C.A.R.Civil Air Regulations
D.L.V.Deutsch Luftsport Vcrband
N.A.A.National Aeronautic Association
N.A.C.A.National Advisory Committee for Aeronautics
S.S.A.Soaring Society of America
INDEX
Aachen,7
Abbreviations, 218
ABC sailplane, 36
Aerodynamics, 15-32
Aileron, 54-56
Air density and lift, 11 j-nG
Air masses, 138-140
Air pressure, 114
Airplane tow, 10, 106-110
Airspeed indicator, 145-146, 193
Airwaxs maps, 181
Albatross sailplane, 37, 191
Altimeter, 146-148. 162
Altitude soaring, 192-195
Altocumuli clouds, 137
Amicale du Vol a Voile Frangais, 216
Archdeacon, Ernest, 5
Artificial horizon, 160
Ascending current, 116
Askania instruments, 148
Aspect ratio, 20
Atmosphere, temp, of, 111-114
Auto tow, 5, 10, 84-95
A very, William, 4
Ball bank indicator, 148
Barbot, 7
Barnaby, Ralph, 10
Barogram, 149, 184, 193
Barograph, 162-163
Barringer, Lewin B., 12, 14, 209
Barstow, Jack, 10
Bibliography, 216
Biskra, 7, 8
Blasting cap, electric, 91-92
Bodecker, A., 13
Bowlus Baby Albatross, 36
Bowlus-du Pont sailplanes, 37, 38, 191
Bowlus, Hawley, 10, 190
Brautigam, 12
Brown, Woodbridge P., 14, 209
Brown-Woodruff winch, 99
Capillary leak variometer, 148, 153-154158
Cayley, Sir George, i
Chanute, Octave, 2, 3, 4, 5, 9, 33
Civil Aeronautics Administration, 40,
106, 181, 223
Civil Air Regulations, 106, 201
Civil certificates, 201
Clinometer, 160, 183
Clock, 160
Cloud streets, 136, 137
Clubs, 210-215
American, 210-214
British, 214-215
Cobb-Slater variometer, 149-150, 158
Cocke, Lt. William, 11
Cockpit, 65-67
Cold front, 140-143
Colvin, Charles H., 145
Compass, 160
Condensation, 129-132
Construction, design and, 40-76, 79-83
Control systems, 58-61, 67-69
Corcoran, Stanley, 191, 209
Cross-country flight, 177, 180, 181, 183,
184, 189, 192
Cumulonimbus, 134
Cumulus clouds, 132-136, 137, 142, 183,
193-194
Current, ascending, 116
Currents, slope, 125-129
Darmstadt, 5
Decker, Chester, 209
Dent, Davis, 154
Deschamps, 8
Design and construction, 40-76, 79-83
Design, detail, 61-76
Dihedral, 27
Diiectional gyro, 160
Distance soaring, 177-191
Dittmar, Heini, 11, 12
Diurnal variations, 118-120
Drag, 18
Dreschel, Walter, 13
du Pont, E. Paul, 99
du Pont, Richard C., 12, 209
du Pont winch, 98
Laton, Warren E., 11, 191
Elmira, 10, 14, 120, 125, 183
Emerik, V., 13
Ksso Touring Service, 181
Etrich, Igo, 5
Falcon sailplane, 191
Feiber barograph, 163
Ferber, Ferdinand, 2, 5
Fin, 27
Fittings, 72-74
Flight, gliding, 15-26
219
220 INDEX
Flight training, 164-174
Flinch, Bernhard, 13
Flugsport, 6, 216
Frankfort, 176
Franklin, R. E., 10
Franklin utility glider, 10, 35, 104
Front, cold, 140-143, 190
polar, 138-140, 190
Fuselage, 43-47, 56
Georgii, Prof., 9
Glider, transport, 196-199
troops, 196-197
Gliding flight, 15-26
Glossary, 217-218
Goal flight, 12, 185, 191-192
"Golden C," 194, 210
"Golden C" pilots, 209
Gradiometer, temperature, 161-162
Groenhoff, Giinther, 10, 12
Gutermuth, Hans, 6
Gyro, directional, 160
Hang glider, 2, 3, 33, 85
Hannover, 7
Hart, Friedrich, 7
Hawks, Frank, 10
Heinemann, 12
Henzen, 7
Herring, A. M., 2, 3
Hesselbach, Peter, 9
High performance sailplane, 36-39, 177
Hirth, Wolf, 10, 125, 188
Horn variometer, 150-152, 158
Humidity, 129-131
Hygrometer, 162
Ilchenko, V. M., 13
Instrument flying, 193
Instruments, 145-163
Intermediate sailplane, 36, 37, 177
ISTUS, 9, 13
Itford Hill, 7
Kantrowitz, Arthur, 154
Kartasheff, I., 13
Kegel, Max, 8, 9
Kitty Hawk, 4, 5, 6
Klemperer, Wolfgang, 7, 10
Klepikova, O., 13
Kollsman instruments, 146, 147, 148,
149
Kronfeld, Robert, 9, 10
Landing gears, 69-72
Lange, Karl O., 100, 111
Lapse rates, 113, 116-118, 119-121, 131132
Launching methods, 84-110
LeBris, Captain, i
Leonard, Parker, 96, 209
Licenses, 201
Lift coefficient, 17
Lilienthal, Otto, 2, 3, 5, 33
Loads and stresses, 40-43

Low pressure area, 140
McConnell, Robert, 152
Maintenance, 76-79
Maneyrol, 7
Martens, Arthur, 7
Meeker winch, 97
Mehlhose, Emerson, 209
Merboth, Warren, 209
Meteorology, soaring, 111-144
Meter, auto-pulley tow, 95
Minimoa sailplane, 38, 75, 190
M.I.T. winch, 99-100
Moazagotl, 143-144
Modlibowska, Wanda, 13
Montgomery, J. J., 5, 33
Mouillard, L. P., i
National Aeronautic Ass'n, 190
Nehring, Johannes, 8, 9
Nessler, Eric, 186
North Conway, 193
Oeltscher, R., 12
Olympic sailplane, 79-83
O'Meara, J. K., 10, 12, 209
Parachute, 181
Performance, 28-32
Pilcher, Percy S., 2
Pioneer instruments, 148
Point Loma, 10
Polar front, 138-140, 190
Primary glider, 33-34
Primary soaring instruction, 172-174
Purdue Glider Club winch, 97
Raspet, August, 145
Rastorgueff, Victor, 13
Records, 202-208
Reitsch, Hanna, 11
Release, 62-64
explosive, 91-93
Reynold's Number, 23
Rhon Mountains, 6, 7, 12, 13
Rhon-Rossitten Gesellschaft, 9
Rhonsperber sailplane, 38
Riedel, Peter, 10, 12, 134
Robinson, John, 186, 209
Ross, Harland C., 209
Ross "Ibis" sailplane, 37, 38
Rossitten, 8
Rotter, Ludwig, 12
Sailplane and glider, 216
Samolet, 216
Savtzov, P., 13
Scheurer, Gustave, 96, 97
Scheurer winch, 97
Schmidt, Kurt, 11
Schools, United States, 210-214
Schultz, Arthur B., 36
Schulz, Ferdinand, 8
INDEX 221
Schweizer, Ernest, 15, 40
Schweizer, Paul, 15, 40
Schweizer sailplanes, 39
Secondary glider, 35-36, 177
Shock-cord launching, 7, 42, 102-106
"Silver C" license, 175, 181, 190, 192
"Silver C" pilots, 209
Sinking speed, 16
Slope currents, 125-129
Slope soaring, 172-173, 175, 177
Soaring flight, 16, 178
Soaring Magazine, 216
Soaring Society of America, 11, 190,201,
218
Soaring technique, 175-195
altitude, 192-195
distance, 177-191
goal flight, 191-192
Spin, 180-181
Stability, -26-28
Stall, 22, 23, 180-181
Stanley, Robert, 14, 209
Steinhoff, 12
Steinmetz, Charles P., 3
Stevens-Franklin glider, 35
Stoughton, Milton, 15
Sweet, Floyd, 209
Tail surfaces, 56-58
Taylor, George F., 138
Temperature gradiometer, 161-162
Thermal soaring, 178
Thermals, 121-125
evening, 125
Thermometer, 161
Total energy variometer, 154-155
Towing releases and hooks, 62-64
Towlines, 85-89, 100
Trailers, 74-76
Training, flight, 164-174
Transport gliders, 196-199
Troops, glider, 196-197
Turn indicator, 159-160, 178, 193
Turns, 178, 179-180, 192
Ursinus, Oscar, 6
Utility glider, 35-36
Vampyr sailplane, 7, 8
Variometer, 148-159, 177, 193
capillary leak, 148, 153-154, 158
Cobb-Slater, 149-150, 158
electrical, 152-153, 158
Horn, 150-152, 158
Kollsman, 148, 149
selection of, 155-159
total energy, 154-155
Vauville, 7
Voisin, Gabriel, 5
Waco primary glider, 34
Warren Eaton Site, 14, 182
Wasserkuppe, 5, 6, 9, 10, 12,
194
Weiss, Jose, 5
Wichita Falls, 14
Wightman, Henry, 209
Winch launching, 92-93, 95-102
Winch tow, 95-102
Wings, 47-54 56
Wright, Orville, 2, 3, 4, 5, 6, 7, 9
Wright, Wilbur, 2, 3, 4 5 7 9
Zander, K. H., 13
Ziller, Erwin, 13
Zook, Elmer, 209
13,

Flight Without Power

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Flight Without Power

Uploaded by

Copyright:

Available Formats

WITHOUT

PILOT'S VIEW WHEN TAKING OFF AT ELMIRA

PITMAN PUBLISHING CORPORATION

All rights reserved. No part of tfiis book

Revised Edition, 1942

ASSOCIATED COM TAN IKS

SIR ISAAC PITMAN & SONS, LTD.

PRINTED IN Till', UNlll'.D SI A IKS OF AMERICA

DEDICATED TO THE MEMORY

During the two years after my first taste of motorlcss flying in

Director for The Soaring Society of America and designer of one

FOREWORD TO THE REVISED EDITION

FOREWORD TO THE REVISED EDITION

HISTORY OF MOTORLESS FLIGHT, by N. H. RandersPehrson

AERODYNAMICS, by Paul Schweizer, Assisted by Milton

TYPES OF GLIDERS, by Lewin B. Barringer ....

DESIGN, CONSTRUCTION AND MAINTENANCE, by Paul and

LAUNCHING METHODS, by Lewin B. Barringer

SOARING METEOROLOGY, by Karl O. Lange .

INSTRUMENTS, by Charles H. Colvin and August Raspet

FLIGHT TRAINING, by Lewin B. Barringer

SOARING TECHNIQUE, by Lewin B. Barringer .

TRANSPORT GLIDERS, by Lewin B. Barringer .

APPENDIX. Gliding and Soaring LicensesRecordsLists of

HISTORY OF MOTORLESS FLIGHT

I HE EARLY HISTORY OF AVIATION is a history of failures